Purpose
Diabetic retinopathy (DR) affects 28.5% of people with diabetes and is the leading cause of new blindness among adults aged 20-74 years. Clinical trials of ranibizumab (RBZ), aflibercept (AFB), and bevacizumab (BVZ) for the treatment of diabetic macular edema evaluated DR severity as secondary and exploratory outcomes. The purpose of this exploratory analysis was to facilitate cross-trial comparisons of anti-vascular endothelial growth factor A (anti-VEGF) agents for DR.
Methods
Study designs, primary endpoints, inclusion/exclusion criteria, baseline (BL) characteristics, and DR outcomes were compared for DRCR.net Protocol-I (RBZ), RIDE/RISE (RBZ), VIVID/VISTA (AFB) and BOLT (BVZ). Fundus photographs (FP) were graded on the Early Treatment Diabetic Retinopathy Study DR Severity Scale (DRSS); secondary endpoints included 2- and 3-step changes in DRSS.
Results
At BL, 24-34% of eyes in Protocol-I and RIDE/RISE had mild/moderate proliferative DR (PDR; DRSS 60-65) vs. 1-7% in VIVID/VISTA and BOLT. In VIVID, 25-29% of FPs were not gradable at BL vs 1-6% in all other trials. In year 1, the proportion of anti-VEGF-treated eyes with ≥2-step DRSS improvement was 21% (Protocol-I), 31-34% (RIDE/RISE), 29-34% (VISTA), 28-33% (VIVID), and 7.5% (BOLT). In contrast, fewer sham/laser-treated eyes improved by ≥2-steps (6.9%, 2.5%, 14.3%, 7.5%, 2.9%, respectively). Worsening of ≥2-steps in year 1 was observed in 2.0% (Protocol-I), 0.4% (RIDE/RISE), 4.6-7.1% (VISTA), 1.2-2.4% (VIVID), and 0% (BOLT) of anti-VEGF-treated eyes vs 7.7%, 7.1%, 14.3%, 8.8%, and 0% of sham/laser-treated eyes, respectively. In year 2, ≥2-step improvements were maintained in RIDE/RISE (36-37%), VIVID (29-33%), VISTA (40%), and BOLT (5.7%). In RIDE/RISE, 2- and 3-step DRSS improvements were seen as early as 6 months.
Conclusions
Inhibition of VEGF can reverse vision-threatening DR (Figure). Across trials, notable BL differences were identified in PDR presence and the proportion of gradable FPs. Overall, these results suggest that inhibition of VEGF is effective in reversing the course of DR and may constitute a new treatment paradigm. Findings should be interpreted with caution given the limitations of cross-trial comparisons.