Impact of aflibercept on retinal nonperfusion in DME

This article was reviewed by Charles C. Wkyoff, MD, PhD

The core results of the VISTA DME phase 3 study (NCT01363440) among eyes with center-involved diabetic macular edema (DME) causing vision loss found that aflibercept (Eylea, Regeneron Pharmaceuticals) treatment resulted in improved visual and anatomic outcomes over 2 years compared with laser treatment.

In a new analysis, investigators studied the impact of aflibercept on quantified areas of retinal nonperfusion among patients in VISTA.

Related: VISTA subanalysis: Macular RNP and clinical outcomes

In a previous analysis of the study,¹ aflibercept was associated with less worsening of the retinal nonperfusion using a quadrant-level analysis compared with sham treatment.

In addition, in the 3-year VISTA and VIVID-DME (NCT01331681) phase 3 studies of DME, intravitreal aflibercept injections (IAI) were compared with the application of macular laser over 2 years.

The IAI administered every 4 or 8 weeks resulted in significant improvements in vision and anatomic outcomes compared with laser.

Post hoc analysis
Charles C. Wykoff, MD, PhD, and several colleagues conducted a post hoc analysis to expand on previous research to gain a better understanding of the changes in retinal nonperfusion by quantifying the areas of macular nonperfusion at baseline and through week 100 of the study.

Related: Turning to new frontier in micropulse laser treatment for retinal disorders

They also assessed the relationship between the changes at baseline in the area of retinal nonperfusion and the best-corrected visual acuity (BCVA), central subfield thickness (CST), the Diabetic Retinopathy Severity Score (DRSS), and the frequency of development of proliferative diabetic retinopathy (PDR).

The study included patients with macular areas of retinal nonperfusion at baseline as identified by the prior quadrant-level analysis.

In the current work, areas of nonperfusion were quantified by a masked reading center in mm2 using fluorescein angiography images at baseline and weeks 24, 52, and 100. Of the 466 patients in the VISTA study, 178 were included in this analysis.

The patients had been randomized to treatment with IAI 2 mg every 4 weeks (60 patients), 2 mg every 8 weeks (55 patients), or laser (63 patients).

The 3 groups were very similar in the patient demographics and disease status; the mean area of retinal nonperfusion at baseline was relatively small at 1.6 mm2.

Related: Clinical trial outcomes for macular edema don’t translate to ‘real-world’ settings

Study findings
Wykoff, an associate professor of clinical ophthalmology at Houston Methodist Hospital in Texas, reported the results on behalf of the VISTA study investigators.

The results showed that there were no significant differences in the changes in the area of retinal nonperfusion through week 100 among the treatment groups.

According to Wykoff, within the treatment groups among both aflibercept-treated populations there was a statistically significant decrease in the retinal nonperfusion area from baseline to week 100 (P = .0052 and P = .0011, for IAI administered every 8 and 4 weeks, respectively), whereas among the laser-treated patients, there was a numerical decrease in retinal nonperfusion from baseline to week 100 that was not statistically significant (P = .2255).

Regarding BCVA, the changes in the size of the area of nonperfusion resulting from the 2 doses of IAI were negatively correlated with increases in vision (P = .0230 and P = .0045, for the treatments administered every 8 and 4 weeks, respectively) and positively correlated with decreases in central retinal thickness (P = .0288 and P = .0002, respectively) but were not found to be correlated with changes in the DRSS.

Related: Bowtie artifacts: Clinical biomarkers for center-Involving DME?

Development of PDR
Wykoff reported that 2.8% of eyes treated with both doses of IAI and 14% of eyes treated with laser had a PDR event through week 100.

Such an event was defined as development of PDR, the application of panretinal photocoagulation, or performance of vitrectomy.

PDR events were found to occur more frequently in eyes with larger areas of nonperfusion at baseline.

The events occurred, according to Wykoff, 2.1 times and 9.2 times more frequently among eyes with baseline RNP areas of 0.615 to 1.255 mm2 and greater than 1.255 mm2 compared with eyes with 0.615 mm2 or smaller areas of RNP, respectively.

Related: Micropulse laser therapy a safe treatment for retinal disorders

The larger areas were more likely to have a PDR event earlier in the course of the trial, he pointed out.

When combining all patients regardless of randomized arm, the extent of baseline retinal nonperfusion was found to be associated with an increased risk of development of a PDR [event] as has been seen in other data sets, including RIDE/RISE, particularly in the patients treated with laser,” he concluded.

Read more by Lynda Charters

--

Charles C. Wykoff, MD, PhD
e:ccwmd@houstonretina.com
Wykoff is a consultant to Regeneron Pharmaceuticals, the company funding the current study.

----

Reference
1. Wykoff CC, Shah C, Dhoot D, et al. Longitudinal retinal perfusion status in eyes with diabetic macular edema receiving intravitreal aflibercept or laser in VISTA study. Ophthalmology. 2019;126(8):1171-1180. doi:10.1016/j.ophtha.2019.03.040