The year 2018 is important to us for further clarifying the role of intravitreal injections of anti-VEGF agents by gathering 5-year data in three of the most common sight-threatening retinal diseases, including neovascular age-related macular degeneration (nAMD), proliferative diabetic retinopathy, and centrally involved diabetic macular edema. The 5-year long-awaited results of intravitreal anti-VEGF treatment (AVT) have been obtained and analyzed. These studies have highlighted the pivotal role as well as limitations of AVT in these three diseases.

First, a cohort study within a randomized clinical trial (Comparison of Age-related macular degeneration Treatments Trials; CATT) focused on the association of individual morphological features of nAMD on visual acuity (VA) after 5-years of AVT.1 The study has shown that adverse OCT features were associated with a worsening in VA from year 2 to 5, including the expanded area of choroidal neovascularization, expanded area of geographic atrophy, new foveal geographic atrophy, new foveal scar, new foveal choroidal neovascularization, new subretinal hyperreflective material within the center 1 mm, new foveal intraretinal fluid, and new retinal thinning. On the other hand, this study also pointed out that, at year 5, the foveal subretinal pigment epithelial fluid and foveal subretinal fluid are associated with improved VA. These observations are helpful for retina physicians to understand the correlation between structural and functional findings and to know when persistent AVT should stop during follow-up of nAMD patients. Most importantly, in this study, the overall evolved OCT findings after AVT are associated with worsening VA to the level below the baseline by year 5. It indicates that multiple factors, in addition to VEGF, are involved in the pathogenesis of nAMD. New therapies other than AVT are required for better management of nAMD. This is a study with high impact on the clinical management of nAMD.

Second, the 5-year results of a multicenter randomized clinical trial comparing the outcomes of prompt panretinal photocoagulation (PRP) with those of long-term intravitreal ranibizumab for treatment of proliferative diabetic retinopathy are available. This study is a continuation of the 2-year report of the Protocol S study conducted by DRCR.net.2 These data show that both treatments are safe, with few serious complications such as neovascular glaucoma. There is no significant difference in VA outcome between two treatment groups. It is important that the more peripheral visual field loss from baseline in the PRP group than in the AVT group at the 2-year Protocol S report diminished in a 5-year subgroup analysis, suggesting that AVT leads to more rapid visual field loss from years 2 through 5. As expected, diabetic macular edema in the AVT group continued to be less frequent than in the PRP group from years 2 through 5. This study revealed a critical point that PRP and AVT are comparable therapies for patients with proliferative diabetic retinopathy in 5-year follow-up. The general understanding is that PRP leads to a more permanent therapeutic outcome than AVT for proliferative diabetic retinopathy. It would be meaningful to study in a real-world setting whether the discontinuation of a long-term AVT is riskier to the visual outcome of proliferative diabetic retinopathy patients beyond 5 years.

The third story is a continuation of Protocol I by DRCR.net from year 1 to 5.3 Protocol I originally showed the superior effect of AVT (ranibizumab) on VA and macular thickening compared with laser alone in year 1. The 2- and 3-year results of the same study confirmed the superior VA achieved through AVT. Meanwhile, the improvement of diabetic retinopathy severity (DRS) during the course of AVT has been observed from year 1 to 5. Impressively, in a subgroup analysis the patients with moderate/severe nonproliferative diabetic retinopathy at baseline had the highest improvement rate. The change of DRS in this group has drawn great attention because these diabetic eyes have a higher tendency toward progressing than do eyes with proliferative diabetic retinopathy. Another critical datum is that improvement in DRS after AVT from year 1 to 5 persisted despite the reduced frequency of the injections over the course of the study, suggesting a sustainable benefit on DRS with AVT in both nonproliferative and proliferative diabetic retinopathy. However, the DRS in 20% of patients with DRS improvement at 1 year reported by Protocol I eventually worsened relative to baseline. Therefore, in order to optimize AVT for diabetic retinopathy in general, the underlying mechanisms by which AVT is beneficial to DRS improvement in some subgroups but not others need further study.