Abstract
Purpose :
VISTA and VIVID showed that patients with center involving diabetic macular edema (DME) treated with intravitreal aflibercept injection (IAI) achieved significantly greater improvements in visual and anatomic outcomes than those treated with thermal laser photocoagulation. This post hoc analysis was performed to assess patients who developed vision-threatening complications (VTCs).
Methods :
VISTA and VIVID were two similarly designed phase 3 studies that treated patients with center involving DME with either IAI 2 mg every 4 weeks (2q4), IAI 2 mg every 8 weeks following 5 monthly doses (2q8), or macular thermal laser photocoagulation (laser control). This post hoc analysis evaluated the incidence of VTCs consisting of proliferative diabetic retinopathy [PDR], panretinal photocoagulation [PRP], or vitrectomy through week 100 in 235 laser control patients and 475 combined IAI patients in the integrated VISTA and VIVID dataset who had a gradable Diabetic Retinopathy Severity Scale (DRSS) score and no PDR at baseline. For patients who received rescue treatment, data were censored from the time rescue treatment was given.
Results :
The proportion of patients who developed a VTC through week 100 was 11.1% vs 4.4% in the laser control vs. IAI group, with an adjusted difference of -6.7% (97.5% confidence interval: -11.7, -1.6; p=0.0008). Specifically, the incidence of development of PDR was 9.8% in the laser control vs. 2.5% in the IAI group. In both treatment groups, all VTCs occurred in patients who had a baseline DRSS score of 43, 47, or 53. Patients with baseline DRSS scores of 10, 20, or 35 did not develop VTCs during the study period.
Conclusions :
Patients with moderate to severe nonproliferative diabetic retinopathy who were treated with laser control experienced a VTC considerably more frequently than patients treated with IAI in the VISTA and VIVID trials.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.