麻豆传媒

Longer-term follow-up of a randomized trial confirmed that prophylactic anti-VEGF therapy with aflibercept (Eylea) slows progression of diabetic retinopathy but does not improve visual acuity better than delayed treatment.

After 4 years, 34% of patients randomized to intravitreous aflibercept injections developed proliferative diabetic retinopathy (PDR) or center-involved diabetic macular edema (CI-DME) with vision loss compared with 57% of controls who received sham injections and deferred anti-VEGF treatment (P<0.001), reported Adam Glassman, MS, of the Jaeb Center for Health Research in Tampa, Florida, and colleagues.

But this delay in disease progression failed to translate to a benefit in visual acuity, with a mean change from baseline of -2.7 letters versus -2.4 letters at 4 years for the two groups, respectively (P=0.52), with no subgroups identified suggesting a potential benefit for the prophylactic approach, according to findings published in .

"This study indicates that monitoring patients regularly for vision-threatening diabetes complications and treating eyes only as needed is the best approach," co-author Raj Maturi, MD, of the Indiana University School of Medicine and Retina Partners Midwest in Indianapolis, said in a , which helped fund the study.

Patients in the study were randomly assigned to either 2.0 mg aflibercept (n=200) or sham injections (n=199) at baseline, 1, 2, and 4 months, and then every 4 months over the first 2 years, continuing through 4 years unless nonproliferative diabetic retinopathy (NPDR) improved to mild disease. For either group, additional anti-VEGF injections were administered as needed to any eye that developed a vision-threatening complication, such as macular edema or PDR.

"While the individual risk of complications per injection is low, the risk increases with each additional injection," co-author Jennifer Sun, MD, MPH, of Harvard Medical School in Boston and chair of Diabetes Initiatives for the Diabetic Retinopathy Clinical Research (DRCR) Retina Network, said in a statement. "The anatomic benefit from early anti-VEGF treatment does not result in improved visual acuity, and so it may not be worth the risk and inconvenience to the patient of repeat preventive injections for NPDR."

The results address the clinical conundrum created by the 2-year results of the study, which at the time also showed the mixed findings: an anatomic benefit that did not translate to improved visual acuity.

"Even with such regular aflibercept treatments, the rates of new PDR and CI-DME in the study continued to rise, from 16.3% at 2 years, to 33.9% at 4 years," wrote Tien Yin Wong, MBBS, PhD, of Tsinghua University in Beijing, and colleagues, writing in an . They added that the hazard ratios for the preventive effects of prophylactic aflibercept became less favorable over time, suggesting that anti-VEGF therapy for NPDR "may have diminishing returns over time."

Asked for his perspective, Rajendra Apte, MD, PhD, of Washington University in St. Louis School of Medicine, said that monitoring patients with diabetes for retinopathy, as recommended by the American Academy of Ophthalmology, remains critical.

In the study, he noted, 40% of patients assigned to sham ultimately initiated aflibercept during the study period for treating PDR or CI-DME and received a mean of 8.7 injections in years 1 to 2 and a mean of 5.6 injections in years 3 to 4.

"This highlights [the importance of careful monitoring] based on the type of diabetes, glucose control, comorbidities and risk factors, and the level of baseline retinopathy, among other factors," Apte told 麻豆传媒.

"As the authors point out, 80% of participants randomized to aflibercept did need at least one injection in year 4, and approximately 20% needed more," he added. "This suggests that in order to maintain the gains made after initiating therapy, chronic injections may be necessary, but how many and for how long after the 4-year period is currently not known."

The DRCR Protocol W study was conducted at 64 clinical sites in the U.S. and Canada from 2016 to 2018. Included were 328 participants (total of 399 eligible eyes) with moderate or severe NPDR, defined as scoring 43-53 on the Early Treatment Diabetic Retinopathy Study () severity scale, which ranges from 0 to 100 (best to worst). The average age of participants was 56 years, and 42% were female. Overall, 45% were white, 32% were Hispanic, 15% were Black, and 5% were Asian.

At 4 years, the probability of developing PDR was 27.9% for eyes in the aflibercept group compared with 49.0% for those in the sham group (adjusted HR 0.42, 97.5% CI 0.29-0.61), while the probability of developing CI-DME with vision loss was 11.3% versus 19.1%, respectively (adjusted HR 0.51, 97.5% CI, 0.27-0.97).

Antiplatelet Trialists' Collaboration cardiovascular/cerebrovascular event rates were 10% in bilateral participants, 11% in unilateral aflibercept participants, and 8% in unilateral sham participants.

Study limitations, the team said, included that participant retention was approximately 75% at 4 years, that measures of visual function other than best-corrected central visual acuity were not obtained, and that other anti-VEGF agents or re-treatment algorithm may produce different results.

Comment