Creation of Retinal Vein Occlusion Model in Cynomolgus Monkeys and Determination of its Pathological Features

Abstract

Objective: A retinal vein occlusion (RVO) is a relatively common retinal vascular disorder, especially in the elderly. Many experiments have been performed on patients with an RVO but performing any type of experiments and especially longitudinal experiments on humans is difficult, if not impossible, on ethical grounds. Therefore, we have created a retinal vein occlusion (RVO) model by laser irradiation of cynomolgus monkeys after intravenous injection of rose bengal. We evaluated the pathological changes of the retina, and the effects of ranibizumab, an anti-vascular endothelial growth factor (VEGF) antibody, on the characteristics of the RVO. Methods: The integrity of the vascular system was evaluated by fluorescein angiography (FA), and the retinal thickness and volume were determined by optical coherence tomography (OCT). The cytokines and growth factors in the aqueous humour were identified by multiplex profiling. Results: Our results showed that ranibizumab decreased the degree of vascular leakage and retinal edema at 1-3 days (acute phase) and 3-7 days (subacute phase), and suppressed foveal thinning at 28-42 days (chronic phase) after the laser irradiation. Ranibizumab also decreased the area of the foveal avascular zone, and the area was negatively and significantly correlated with the thickness of the ganglion cell layer (GCL) complex. Furthermore, ranibizumab reduced the increased expression of VEGF in the aqueous humor, but did not affect the expressions of interleukin-6 (IL-6), monocyte chemotactic protein-1 (MCP-1), angiopoietin-1 (ANG-1), or angiopoietin-2 (ANG-2). These findings suggest that ranibizumab attenuates the retinal edema and subsequent retinal atrophy in part by neutralizing VEGF. However, other cytokines and growth factors were also affected by the ranibizumab, which suggests that not only VEGF but also other unidentified agents might play a role in the pathogenesis of the RVO. Conclusion: We have created a non-human primate RVO model, which resembles the clinical RVO pathology. In this model, an injection of ranibizumab leads to a reduction in vascular leakage and the retinal thickness and volume by blocking the expression of VEGF. Our model might be useful for investigating the pathological mechanisms of RVOs and explore new therapeutic agents for RVO.