Mixed Results From Retinopathy Prevention Trial

Prophylactic anti-VEGF therapy with aflibercept (Eylea) slowed the progression of diabetic eye disease but did not preserve visual acuity any better than deferred treatment, a randomized trial showed.

After 2 years of follow-up, patients randomized to aflibercept had a 16.3% probability of developing center-involved diabetic macular edema (CI-DME) with vision loss or proliferative diabetic retinopathy (PDR) as compared with 43.5% in the control group. Separate analyses showed that both outcomes were reduced by more than 50% with aflibercept versus deferred anti-VEGF treatment.

Despite the favorable impact of prophylactic aflibercept on diabetic eye disease, the mean change in visual acuity did not differ significantly between treatment groups (-0.9 vs -2.0 letters), Adam R. Glassman, MS, of the Jaeb Center for Health Research in Tampa, Florida, and co-authors reported in JAMA Ophthalmology.

"Additional follow-up is needed to determine whether early treatment leads to visual benefit long term," the researchers wrote. "Eyes receiving treatment for PDR and CI-DME often have suboptimal visual outcomes; therefore, preventing these conditions may reduce vision loss over time."

The mixed results create a conundrum for clinical ophthalmology practice, according to the authors of two accompanying editorials. Do the results make a case for proactive intervention to prevent progression of nonproliferative to proliferative diabetic retinopathy?

"Based on the Protocol W results, the natural history of NPDR [nonproliferative PDR] progression to PDR and CI-DME, and the known risks and benefits of prophylactic aflibercept, my answer is yes," wrote Jennifer I. Lim, MD, of the University of Illinois at Chicago and associate deputy editor of JAMA Ophthalmology.

"In a patient with severe NPDR, it is reasonable to consider anti-VEGF prophylaxis to lower the patient's diabetic retinopathy severity score (DRSS) and potential risk of progression to PDR. For eyes with severe NPDR, patients often harbor comorbidities that may contribute to delays in treatment, loss to follow-up, and worsening of retinopathy," Lim continued.

Data from the Diabetic Retinopathy Clinical Research (DRCR) Network Protocol W and from the PANORAMA trial (which showed improvement in DRSS and reductions in anterior segment neovascularization, PDR, and CI-DME) make prophylactic anti-VEGF treatment a reasonable consideration for patients with moderate or severe NPDR, she noted. Moreover, increasing DRSS has a significant association with worsening quality of life.

"In this group with a high risk of progression to CI-DME or PDR, treatment reduces the risk of progression and contributes to reduction of morbidities that visual acuity alone cannot measure," Lim concluded.

The authors of the second editorial suggested a wait-and-see approach to the Protocol W results, highlighting results from previous trials of anti-VEGF therapy. The CLARITY trials with aflibercept and the DRCR Protocol S with ranibizumab (Lucentis) showed that anti-VEGF treatment can induce regression of neovascularization in eyes not previously treated with pan-retinal photocoagulation (PRP), but the regression did not correlate with improved vision, noted Rajendra S. Apte, MD, PhD, and Christopher K. Hwang, MD, PhD, both of Washington University in St. Louis.

A preplanned cost-effectiveness analysis of Protocol W suggested that intravitreal treatment with ranibizumab was not cost-effective compared with PRP at 2 or 5 years, those editorialists said. Moreover, the morbidity and costs associated with a small risk of endophthalmitis with intravitreal anti-VEGF therapy should not be overlooked when considering potential prophylactic treatment of a large patient population.

"Four-year follow-up for Protocol W is scheduled to be completed in 2022," Apte and Hwang noted. "Waiting no more than 2 years from now for 4-year visual acuity outcomes may be warranted for eyes with moderate to severe NPDR, as these eyes are monitored safely in Protocol W. The 4-year follow-up could also shed some light on the natural history of aflibercept-treated eyes with mild NPDR [during 2 years of nontreatment] and provide some reference for comparison with anti-VEGF-naive eyes with mild NPDR."

For DRCR Protocol W, investigators at 64 sites in the U.S. and Canada enrolled, randomized, and treated 328 patients (total of 399 eyes) with moderate or severe NPDR (Early Treatment Diabetic Retinopathy Study [ETDRS] severity scale 43-53) to intravitreal injections of aflibercept or sham injections. Injections occurred at baseline, and 1, 2, and 4 months, and then every 4 months for the first 2 years.

During years 2-4, treatment was deferred so long as the eye had mild NPDR or better status. For patients in either group, aflibercept was administered if CI-DME with vision loss or high-risk PDR developed. The primary endpoint was the proportion of patients with CI-DME with vision loss or high-risk PDR after 2 years.

The 27% absolute difference in the primary endpoint represented a 68% reduction in the hazard ratio in favor of prophylactic aflibercept (97.5% CI 0.21-0.50, P<0.001), the researchers reported. The 2-year probability of PDR was 13.5% with aflibercept and 33.2% for the control arm, and the probability of CI-DME with vision loss was 4.1% versus 14.8%.

Analysis of change in visual acuity, a secondary outcome, yielded a modest, nonsignificant trend in favor of aflibercept, Glassman and co-authors said. After adjustment, the between-group difference shrank to 0.5 ETDRS letters (97.5% CI -1.0 to 1.9, P=0.47).