Purchase this article with an account.
Geeta Lalchandani, Na Lu, Ivaylo Stoilov; Diabetic macular edema outcomes with anti-VEGF treatments: comparison of randomized controlled clinical studies. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):1766.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Comparing outcomes across trials of ranibizumab (RBZ), aflibercept (AFL), and bevacizumab (BVZ) for the treatment of diabetic macular edema is challenging due to differences in study design and implementation. This exploratory analysis compared BCVA outcomes while accounting for differences in baseline (BL) characteristics.
Study designs, primary endpoints, inclusion/exclusion criteria, BL characteristics, and BCVA outcomes were compared for DRCR.net Protocol-I (RBZ), RISE/RIDE (RBZ), VIVID/VISTA (AFL) and BOLT (BVZ). Inclusion criteria were similar except for BL BCVA: Protocol-I and RESTORE enrolled patients with BCVA up to 20/32 (Snellen equivalent), whereas the others capped vision at 20/40. To compare across trials, a Protocol-I subgroup analysis was conducted, including only study eyes with BL BCVA 20/40 or worse (n=581, 68%).
Patients with a previous history of laser treatment are reported as: Protocol-I: 54%, RIDE: 58-62%, RISE: 69-72%, VIVID: 65%, and VISTA: 53%; 12-19%, 26-29%, 31-40%, 6%, 24% had intravitreal steroids, respectively. Prior anti-VEGF was received by patients in Protocol-I: 11-13%, VIVID: 9%, and VISTA: 43%. RESTORE, BOLT did not report prior treatment. BL HbA1c levels were 7.4-7.9% across all trials. Across all studies, most BCVA gains occurred during Year 1 (Yr1). At Yr1, patients gained a mean +10-12 letters from BL in RISE/RIDE, which was clinically comparable to VIVID/VISTA (+11-13 letters), and higher than BOLT (+5.6 letters). In RESTORE, patients gained +6-7 letters from BL at Yr1. In the Protocol-I subanalysis, mean BCVA gains from BL were +11 letters in both the prompt and deferred laser groups, and were greater than in the all-comers population (+9 letters).
When adjusted for BL BCVA, BCVA gains from BL appeared to be comparable across studies between RBZ and AFL given monthly or less-than-monthly, and were less with BVZ. Comparatively smaller BCVA gains in the Protocol-I and RESTORE general populations could be related to a ceiling effect on vision gains due to better-seeing eyes at enrollment. Differences in prior treatment may also have affected outcomes. Findings should be interpreted with caution given the limitations of cross-trial comparisons.
This PDF is available to Subscribers Only