Impact of baseline retinal nonperfusion on best corrected visual acuity in eyes with central retinal vein occlusion: post hoc analyses of COPERNICUS and GALILEO

Abstract

Purpose : Retinal nonperfusion may be an important factor impacting the efficacy of anti-VEGF agents in macular edema (ME) due to central retinal vein occlusion (CRVO). This post hoc analysis evaluates its role in COPERNICUS and GALILEO.

Methods : Patients with ME due to CRVO were randomized (3:2) to receive IAI 2 mg or sham monthly for 24 weeks. From Week 24 (W24) until W52, all IAI–treated patients in both studies and the sham-treated patients in COPERNICUS were eligible to receive IAI based on visual and anatomical criteria. In these post hoc analyses, eyes were classified as having any macular retinal nonperfusion (MNP) determined by fluorescein angiography (FA) within the ETDRS field centered on the macula and as meeting the historic Central Vein Occlusion Study (CVOS) criteria of “ischemia” (CVOS “ischemic” Group [≥10 DA of nonperfusion on FA anywhere in the standard fundus seven subfields]). The association between eyes with any MNP and eyes with ≥10 DA of nonperfusion at BL were assessed by the kappa coefficient. Data from COPERNICUS and GALILEO were integrated.

Results : At BL, 51 eyes had ≥10 DA of nonperfusion in 7 standard fields; of these, 48 (94.1%) had MNP and 3 (5.9%) had no MNP. Of the 230 eyes with <10 DA nonperfusion in 7 standard fields at BL, 156 (67.8%) had MNP and 74 (32.2%) had no MNP. Distribution of retinal perfusion status was balanced between the IAI and sham groups. There was low agreement between BL macular perfusion status and BL CVOS “ischemic” criteria (Kappa=0.12). The treatment effects at W52 with respect to mean changes in BCVA from BL were consistent regardless of BL retinal perfusion status in 7 standard fields or MNP status for IAI (+14.2 to 16.7 letters) versus sham (+0.6 to 5.6 letters). By W52, the most frequent ocular serious adverse events were vitreous hemorrhage in COPERNICUS (3.2% of patients) and ME in GALILEO (3.5%). There were 3 Anti-Platelet Trialists’ Collaboration–defined arterial thromboembolic events in COPERNICUS (IAI, n=1; sham, n=2) and none in GALILEO.

Conclusions : In COPERNICUS and GALILEO, which included eyes meeting historic CVOS “ischemic” criteria and eyes with MNP, the treatment benefit of IAI on mean change in BCVA was robust and similar regardless of these BL characteristics.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.