Two-year results from the Protocol W clinical trial of the DRCR Retina Network (formerly the Diabetic Retinopathy Clinical Research Network) suggest that early treatment of pre-existing diabetic eye disease with injections of the anti-VEGF agent aflibercept slow the development of vision-threatening complications of diabetic retinopathy.

At the halfway point of the study, however, the effects of the treatment on visual acuity were similar to those obtained with initially observing eyes and treating only if vision-threatening complications developed.

The researchers note that vision-loss in diabetic retinopathy may develop later in the disease, suggesting that the results at four years will be important for judging the longer-term effects of treatment. The findings were published March 30th, 2021 in JAMA Ophthalmology. “We performed this study to address an important question for patients–-whether preventive treatment with injections of aflibercept should be performed for patients who have diabetic retinopathy and who are at risk for vision loss with the development of more advanced disease,” said senior author Jennifer Sun, MD, MPH, of the Joslin Diabetes Center and chair of diabetes initiatives for the DRCR Retina Network.

“We know that anti-VEGF therapy is very effective in improving vision when diabetic eye complications develop. This study was designed to see if earlier treatment would provide even more favorable long-term results.”

Diabetic retinopathy is a complication of diabetes which can cause blood vessel abnormalities in affected eyes. In the early stages of the complication, termed non-proliferative diabetic retinopathy (NPDR), changes in blood vessels in the retina are visible but frequently do not result in vision loss for the patient.

However, as the condition worsens, typically over many years, individuals with diabetes can develop progressive diabetic retinopathy (PDR; where retinal blood vessels grow abnormally) or diabetic macular edema (DME; where retinal blood vessels leak). In both cases irreversible vision loss, and even blindness can result.

In the study, 329 individuals with NPDR and good vision were randomly assigned at baseline to receive injections into the eye of aflibercept or a sham (pretend) injection at one, two and four months and then every four months after that up to two years. Researchers then monitored vision changes and any progression of diabetic retinopathy at each visit. Regardless of group assignment, individuals received standard anti-VEGF treatment if the condition worsened to DME with vision loss or severe PDR.

The probability in the first two years of the study of developing either PDR or DME with vision loss was 43.5% for individuals who received placebo. This compared to 16.3% in individuals who received aflibercept over the same period. The probability of developing PDR was 13.5% with the treatment and 33.2% with placebo. For DME with vision loss, the probability was 4.1% with aflibercept and 14.8% with placebo.

However, in terms of visual acuity, there was essentially no difference between the groups at two years, suggesting that, at least in the non-proliferative stages of diabetic retinopathy, close observation and appropriate treatment if complications develop are a reasonable approach.

“Protocol W tells us that for eyes with moderate or worse non-proliferative diabetic retinopathy, early treatment with aflibercept is effective at preventing advanced complications of DME with vision loss and PDR,” said Dr. Sun. “However, based on these two-year study results, many physicians and patients may elect to hold off on treatment unless these complications develop, since there does not appear to be a visual benefit from starting injections early.”

The study now continues its follow-up for another two years to answer the question as to whether the two-year anatomic benefits will successfully translate into longer-term visual benefits. According to Dr. Sun, it may be the case that visual decline occurs much later in disease progression and that more study time is needed to test this hypothesis.

“The four-year follow-up data will be critical for understanding the longer-term effects of this preventive approach,” added Sun. “The results after four years will tell us whether the favorable anatomic results at two years translate to longer term superiority of vision outcomes. The bottom line for clinicians at the moment is that regular follow-up and standard treatment if complications develop can result in excellent vision outcomes for patients with diabetic retinopathy. I think we should wait for the four-year results and if at that point we see visual benefits, we will then need to reevaluate the role of early preventive injections in our patients with diabetic retinopathy.”

Other members of the writing committee include Raj Maturi, Adam Glassman, Kristin Josic, Andrew Antoszyk, Barbara Blodi, Lee Jampol, Dennis Marcus, Daniel Martin, Michele Melia, Hani Salehi-Had, Cynthia Stockdale, and Omar Punjabi, on behalf of the DRCR Retina Network.

Funding for the research was provided by the National Institutes of Health, the US Department of Health and Human Services, Regeneron and the DRCR Retina Network. Full funding details and author disclosures are available in the JAMA Opthalmology report.