Larger Choroidal Neovascular Area Predicts Vision Loss Despite Anti-VEGF Therapy

The likelihood of vision loss at 3 months or 1 year despite anti-VEGF therapy increased significantly in patients who had a larger choroidal neovascular (CNV) area at baseline, French investigators reported.

Increasing CNV area had significant associations with both early (within 3 months) and late vision loss ≥5 letters. Presence of subretinal fluid (SRF) at 3 months was associated with ≥5-letter vision loss, and the association strengthened with the presence of both SRF and intraretinal fluid (IRF).

The findings are consistent with a post hoc analysis of the randomized VIEW trial, which also identified CNV area as a factor in vision loss during anti-VEGF treatment for neovascular age-related macular degeneration, reported Laurent Kodjikian, MD, PhD, of Croix-Rousse University Hospital in Lyon, France, and co-authors, in Ophthalmology Retina.

“Looking for predictive factors of vision loss that might determine whether the patient belongs to one or another of these risk groups [for vision loss] could allow for the early identification of a visual prognosis and justify more intensive treatment regimens for certain patients when relevant,” the team wrote. “We found that only baseline CNV area was correlated to the visual trajectory.”

“The similarity of our findings to the post hoc analysis of the VIEW study suggests that baseline CNV area is a predictive factor of vision loss independently of the treatment regimen and anti-VEGF molecule used,” the authors added.

The vision loss observed in the study is consistent with other studies and shows that most patients with neovascular, or wet, AMD do well when treated appropriately, said Sunir Garg, MD, of the Wills Eye Hospital and Thomas Jefferson University in Philadelphia, who was not affiliated with the study.

“However, some eyes, particularly those with larger wet macular degeneration lesions at baseline, are at higher risk of vision loss over time,” Garg, a clinical spokesperson for the American Academy of Ophthalmology, told MedPage Today via email.

Recent advances have improved early recognition of AMD, he continued. A simple at-home test consists of covering one eye and then the other to see whether straight lines become wavy or distorted, a possible indication of wet macular degeneration. Automated at-home screening tests are also available.

“In the near future, we are hopeful that we will have a home optical coherence tomography (OCT) screening system,” said Garg. “OCT is the main image that we use to diagnose and to follow treatment response for eyes with wet macular degeneration. A few companies are working on a version that patients can take in the comfort of their home, and this can detect development of wet macular degeneration hopefully at a very early stage.”

Although development of anti-VEGF therapy has greatly improved outcomes in neovascular AMD, some patients (15-23% in recent studies) have significant vision loss despite treatment, Kodjikian and co-authors noted. In the randomized GEFAL trial comparing bevacizumab (Avastin) and ranibizumab (Lucentis), French investigators found that 9-10% of patients lost at least 15 letters during the first year of treatment.

“Early identification of these patients with worse 1-year outcomes would make it possible to adapt their treatment regimens to prevent vision loss,” the authors stated.

GEFAL investigators retrospectively reviewed 1-year data for the 393 randomized patients. They separated the patients into three categories by change in best-corrected visual acuity (BCVA) at 3 months and 1 year: no vision loss ≥5 letters at either assessment; no vision loss ≥5 letters at 3 months but lost ≥5 letters at 1 year (secondary loss); patients with ≥5 letters at 3 months and 1 year (initial loss).

The study population had a mean age of 79, and women accounted for 67% of the total. Mean baseline BCVA was 55 letters, and mean central macular thickness (CMT) was 357 μm. CNV was purely occult in 44%, and mean total area of CNV was 2.0 disc areas.

The data showed that 225 (57.3%) of patients did not have ≥5 letters vision loss, 109 (27.7%) had secondary vision loss ≥5 letters, and 59 (15%) patients had initial loss ≥5 letters. Final BCVA differed significantly (P<0.0001) between patients with absence of significant vision loss (67 letters) versus those with secondary loss (49.5) and initial loss (45.5).

Baseline total CNV area was significantly greater in patients with secondary or initial loss (2.2 disc area) versus those without vision loss ≥5 letters (1.8, P=0.0412). Neither baseline BCVA nor CMT was associated with vision trajectory.

Presence of SRF at 3 months had a significant association with vision loss (P=0.0286), but presence of IRF did not. The association with initial vision loss increased with presence of both SRF and IRF (P=0.0066). Presence of IRF alone was associated with secondary vision loss.