DME trials show advantage to early aflibercept treatment

Patients treated early with aflibercept (Eylea, Regeneron Pharmaceuticals) for diabetic macular edema (DME) gain more visual acuity than those treated later in the course of the disease, a researcher said.

“If you don’t treat DME soon enough, you leave vision on the table,” reported David M. Brown, MD, a retina specialist at Houston Methodist Hospital’s Blanton Eye Institute and a clinical professor of ophthalmology at Baylor College of Medicine, Houston.

The 148-week results from the VISTA and VIVID trials of aflibercept in DME also show that most patients get good results with injections every 8 weeks, but that some may require more frequent treatment to keep the retina totally dry.

The two double-masked trials took place simultaneously, with VIVID’s 404 patients at 73 centers in Europe, Japan, and Australia, while VISTA’s 461 patients were at 54 centers in the United States. All the patients had clinically significant DME with central involvement and best-corrected visual acuity (BCVA) of 20/40 to 20/320.

The investigators randomly assigned patients to either 2 mg of aflibercept every 4 weeks (Q4), 2 mg of aflibercept every 8 weeks after 5 initial monthly doses (Q8), or laser photocoagulation. The patients receiving photocoagulation also received sham injections, and the patients receiving aflibercept also received sham laser treatment.

The groups were similar in age and severity of disease. The mean ages were 61 to 65 years. HbA1c ranged from 7.6 to 7.9. Duration of diabetes ranged from 14.3 to 17.6 years. Mean BCVA letters ranged from 58.9 to 60.8. Mean central retinal thickness (CRT) ranged from 479 µg to 540 µg.

Differences in demographics

There were some differences in demographics between the 2 trials. In the VIVID trial, about half were female. While in the VIVID trial, two-thirds were female.

In the VIVID trial, the percentage with prior anti-vascular endothelial growth factor (VEGF) treatment ranged from 8% to 15%, while in the VISTA trial it ranged from 63% to 68% as anti-VEGF treatments were more common in the United States than in Europe and Asia.

Beginning at week 24, the patients receiving aflibercept could also receive photocoagulation–and vice versa–as rescue treatments if their disease was worsening. At week 100, the patients who had received photocoagulation could receive aflibercept as needed.

Between 70% and 80% of the patients completed week 148 in the 3 arms of each of the 2 trials,

In the VIVID trial, patients in the photocoagulation group received a mean of 2.6 laser treatments. During the three-year duration of the trial, the number of injections was 32 in the Q4 group and 18.1 in the Q8 group. In the VISTA trial, patients received a mean of 3.8 laser treatments in the photocoagulation arm and 29.6 injections in the Q4 group and 18.1 injections in the Q8 group.

In the VIVID trial, 82% of the patients in the photocoagulation group received either rescue or as-needed aflibercept. In the VISTA trial, 87% received of the photocoagulation group received aflibercept. Fewer than 12% of aflibercept patients received photocoagulation in the 2 trials.

In both aflibercept groups in both trials, patients rapidly gained visual acuity over the first year, reaching a mean improvement of 10 to 12.5 letters BCVA over baseline. Their mean BCVA remained about constant after that, or declined by 1 or 2 letters at week 148.

By contrast, the photocoagulation groups – excluding those patients who also received rescue treatment – got improvements of only 1 to 2 letters, which remained constant through week 148.

When the patients who received rescue treatments were included in the analysis, the VISTA photocoagulation group improved by a mean of 8.2 letters, and the VIVID photocoagulation group improved by a mean of 7 letters. By contrast, including patients from the aflibercept groups who received photocoagulation made little difference in their results.

This result shows that patients are unlikely to gain as much visual acuity if they start treatment years after their diagnosis, Dr. Brown said. “They still don’t do that well,” he added. “Two years is too late to start treatment.”

The mean change in CRT as measured on optical coherence tomography (OCT) paralleled the results in visual acuity. In the VISTA trial, the patients receiving photocoagulation lost 84 µg of thickness at week 100. They were allowed to take aflibercept at that point, and reached an improvement of 110 µg. The patients in the Q4 aflibercept group of the VISTA trial lost 200 µg, and those in the Q8 group lost 190 µg. 

“The Q8 group did great,” said Dr. Brown. “It shows us with this very strong anti-VEGF treatment, the average patient can go to every 8 weeks – or maybe even less, though we didn’t study that.”

However the study did show fluctuation in the CRT of the Q8 patients. “Every 8 weeks did not keep all the patients totally dry,” said Dr. Brown. “In clinical practice, what that meant is that not every patient will get by with Q8 dosing.”

Small number of events

In the VISTA trial, 6.8% of the patients in the aflibercept groups experienced serious ocular adverse events, compared to 4.5% of the patients in the photocoagulation group. Likewise, 47.6% of the aflibercept patients in the VISTA trial experienced serious systemic adverse events, compared to 48.7% of the patients in the photocoagulation group. The results in the VIVID trial were similar. And none of the differences between groups were statistically significant, Dr. Brown said.

“It’s relieving to know that, in this diabetic population, even in those who got bilateral Eyelea did not have significant side effects,” he said.

However, he cautioned that the populations in the studies were healthier on average than those seen in clinical practice. In addition, their participation in the trials may have improved their overall health by increasing their interaction with health care providers.

This 148-week data brings to an end the VISTA and VIVID trials, but Regeneron will continue to monitor some patients.

While answering some of the most important questions about the drug in this population, it leaves a few unanswered, Dr. Brown pointed out.

First, it’s not clear how many injections patients need after 3 to 4 years, or whether some might be able to stop medication altogether. In the open-label extension of the RISE/RIDE trials of ranibizumab for DME, about a quarter of patients did not need injections in their 5^th year of treatment, Dr. Brown said.

Some of this effect may result from patients taking other steps to improve their health, he said.  “When patients get eye trouble, they often realize they need to get serious about their blood sugar control.”

A second lingering question is whether laser photocoagulation provides an additional benefit to patients already being treated with aflibercept.  “Is there any value or is it harmful?” Dr. Brown asked.

The answers will depend on future research.