Aflibercept DME superiority diminishes in second year

Aflibercept remains superior to bevacizumab after 2 years of treating patients with poor baseline visual acuity (VA) caused by diabetic macular edema (DME), but it loses its edge over ranibizumab, researchers said.

“Vision gains from baseline at 2 years were seen in all 3 groups with about half the number of injections, slightly decreased frequency of visits, and decreased amounts of laser in the second year,” said John A. Wells, MD, Palmetto Retina Center, Columbia, South Carolina.

Dr. Wells and his colleagues from the Diabetic Retinopathy Clinical Research Network (DRCR.net) published the findings from the Protocol T trial in Ophthalmology.

The results may help clinicians faced with the difficult choice between the 3 anti-vascular endothelial growth factor (VEGF) drugs approved for DME.

Separately, all 3 drugs have shown impressive results in restoring or protecting vision in patients with DME, but their prices are very different. Based on Medicare allowable charges, the drugs at the doses used in this study cost about $1,960 for aflibercept (Eylea, Regeneron), $1,200 for ranibizumab (Lucentis, Genentech), and about $70 for bevacizumab (Avastin, Genentech) per injection, according to the National Eye Institute.

Recent: Managing silicone oil complications

The DRCR.net investigators enrolled 660 people with DME at 88 sites across the United States. At baseline, the patients had a mean age of 61 years and had type 1 or type 2 diabetes for a mean of 17 years.

The investigators restricted the trial to people with a VA of 20/32 or worse. At enrollment, about half of the participants had 20/32 to 20/40 vision, and the other half had 20/50 or worse vision.

The researchers randomly divided the participants into thirds. One group received 2.0 mg/0.05 mL of aflibercept, one received 1.25 mg/0.05 mL of bevacizumab, and one received 0.3 mg/0.05 mL of ranibizumab.

Clinicians evaluated participants every 4 weeks during the first year and every 4 to 16 weeks during the second year, depending on treatment course. At each visit, they assessed the study eyes for retreatment with the anti-VEGF agent based on visual and optical coherence tomography (OCT) criteria. They administered intravitreal injections of the study drugs until the DME resolved or stabilized.

VA improvement

Additionally, they administered focal/grid laser photocoagulation treatment if DME persisted without continual improvement after 6 months of injections.

Participants did not know which drug they were receiving until the primary results were published in February 2015. At that time, if the investigators considered it to be warranted, the study participant could switch anti-VEGF agents.

After 2 years, 201 patients (90%) remained in the aflibercept group, 185 (85%) in the bevacizumab group, and 218 (88%) in the ranibizumab group. At that point, the aflibercept group had received a median of 15 injections, the bevacizumab had received 16, and the ranibizumab group had received 15.

Sixty-four percent of the bevacizumab group received laser treatment, compared to 41% of the aflibercept group and 52% of the ranibizumab group.

On average, VA improved over the 2 years by 12.8 letters with aflibercept, 10.0 with bevacizumab, and 12.3 with ranibizumab. The difference was statistically significant (p = 0.02) for aflibercept versus bevacizumab, but not for aflibercept versus ranibizumab (p = 0.47) and not for ranibizumab versus bevacizumab (p = 0.11).

Among patients with an initial VA of 20/50 or worse, improvement was 18.1 letters with aflibercept, 13.3 with bevacizumab, and 16.1 with ranibizumab. Once again, this difference was only statistically significant for aflibercept versus bevacizumab (p = 0.02).

More retina: Anti-VEGF induces regression of diabetic retinopathy in high-risk populace

With VA of 20/32 or 20/40, mean improvement at 2 years was 7.8 letters for aflibercept, 6.8 for bevacizumab, and 8.6 for ranibizumab, and none of these differences were statistically significant.

These findings carried forward many of the trends from the previous year. At the 1-year follow-up, when VA was 20/32 to 20/40 at the start of the trial, vision improved on average almost 2 lines in all three treatment groups.

Recent: CATT follow-up explores long-term outcomes with anti-VEGF in AMD

At the end of the first year, for participants whose VA was 20/50 or worse at the start of the trial, aflibercept improved vision on average almost 18.9 letters, bevacizumab 11.8 letters, and ranibizumab 14.2 letters, with a statistically significant difference between aflibercept and bevacizumab (p < 0.001) as well as between aflibercept and ranibizumab (p = 0.003), but not between ranibizumab and bevacizumab (p = 0.21).

Averaging the results over the 2 years, aflibercept maintained its statistically significant superiority over the other 2 drugs.

Central subfield thickness

The mean change in central subfield thickness was greatest with aflibercept, with a drop of 172 µm by the end of the second year. This was better than the drop of 149 µm for ranibizumab and 126 µm for bevacizumab, with the differences between aflibercept and bevacizumab (p < 0.001) and ranibizumab (p = 0.08) statistically significant, but not between ranibizumab and bevacizumab (p = 0.001).

Among patients with a baseline VA of 20/50 or worse, bevacizumab reduced central subfield thickness by 185 µm, which was more than ranibizumab’s 174 µm, though still not as much as aflibercept’s 211 µm.

More: EyeGuru.org seeks to deliver exactly what residents need

In patients with VA of 20/32 to 20/40, on the other hand, bevacizumab only reduced central subfield thickness by 68 µm, versus 125 µm for ranibizumab and 133 µm for aflibercept.

There were no statistically significant differences in the percentages of patients suffering adverse events among the groups. Elevated IOP was the most frequently reported complication.

In his own practice, Dr. Wells said, he still prefers aflibercept for patients with the weakest vision. “I would think clearly baseline vision is important. And if baseline vision were 20/50 or worse, I would choose [aflibercept],” he said. “It’s obviously more effective at 1 year, and even though the difference with ranibizumab is narrowed at the 2-year endpoint, you want to give people better vision as quickly as you can.”

In eyes with better vision, he prefers ranibizumab because it is less expensive than aflibercept and because in these patients, it might reduce edema more than bevacizumab.

More: Is your city is in the nation's top 10 eye-healthy cities?

This article was adapted from Dr. Wells’ presentation during the 2016 meeting of the American Society of Retina Surgeons. Dr. Wells discloses grants and non-financial support from Genentech and Regeneron during the conduct of the study.