Analysis of the Harbor Study data could not definitively answer the question about whether anti-vascular endothelial growth factor (VEGF) therapies influence the development of atrophy in patients with neovascular age-related macular degeneration (AMD). However, investigators did find that the visual acuity (VA) continued to improve in patients treated with monthly and as-needed drug regimens and that certain risk factors can predict development of atrophy, according to Srinivas Sadda, MD.
Los Angeles-Analysis of the Harbor Study data could not definitively answer the question about whether anti-vascular endothelial growth factor (VEGF) therapies influence the development of atrophy in patients with neovascular age-related macular degeneration (AMD). However, investigators did find that the visual acuity (VA) continued to improve in patients treated with monthly and as-needed drug regimens and that certain risk factors can predict development of atrophy, according to Srinivas Sadda, MD.
The course of AMD is well recognized. The pathway of AMD progression proceeds from drusen deposition to atrophy and choroidal neovascularization (CNV) can be an interval event that develops in some patients, Dr. Sadda explained. Before the introduction of anti-VEGF therapy, scarring and vision loss occurred.
With the anti-VEGF drugs, physicians have the opportunity to reverse this process.
“However, we have also recognized that atrophy can still develop in many of these patients,” said Dr. Sadda, Doheny Eye Institute, Los Angeles.
Consideration of factors that lead to atrophy in treated neovascular AMD points to a few culprits.
“Natural progression of the underlying dry AMD may be responsible for development of atrophy,” he said. “Another possibility is that atrophy is associated with the destructive aspects of the exudative CNV disease process, or macular atrophy could actually be associated with anti-VEGF therapy independent of CNV.”
Thus, the challenge is assessing atrophy in CNV, Dr. Sadda said.
He pointed out that there is no consensus methodology by which to make this assessment or even a consensus name, for the purposes of this discussion he referred to it as macular atrophy.
“It can be difficult to identify atrophy with obscuring features from the CNV lesions,” he said. “To ideally study the impact anti-VEGF treatment, atrophy not adjacent to the CNV lesion should be investigated. However, this is a relatively rare event, the lesion can growth over time, and determining if the lesion is not adjacent to the atrophy is difficult to do.
“This may explain some of the differences between the rates of atrophy observed between the CATT and IVAN studies,” Dr. Sadda explained.
Between those two trials, the investigators did not find a consistent effect of the anti-VEGF drug as a risk factor for atrophy. “However,” Dr. Sadda said, “the atrophy rates did appear to be higher in patients treated with mandated monthly injections as opposed to the patients who received as-needed treatment in both studies.”
When considering the results of this retrospective analysis of the Phase III Harbor Study, Dr. Sadda noted that because of post-hoc analyses, the results must be viewed cautiously.
The Harbor study had as-needed arms and monthly treatment arms with different ranibizumab (Lucentis, Genentech) doses (0.5 mg and 2.0 mg).
Masked graders at the reading center reread more than 17,000 images (fluorescein and color fundus photographs) to assess the presence of atrophy, the definitions of which were similar to those used in previous studies.
The Harbor Study tried to look at atrophy that was not adjacent to the CNV lesions, but also considered a more inclusive definition (i.e. all of the atrophy present), according to Dr. Sadda.
The key findings, he explained, included the mean best-corrected VA (BCVA) improved over time in patients with macular atrophy present at baseline who were treated with monthly or as-needed treatments. The presence of macular atrophy at baseline was not associated with an increased proportion of patients losing 15 or more letters of vision. Patients with any concurrent macular atrophy at months 12 and 24 had comparable mean BCVA to patients without macular atrophy at months 12 and 24. The risk factors for atrophy included the presence of cysts in the retina and atrophy in the fellow eye. Interestingly, subretinal fluid was associated with a lower risk of development of atrophy, Dr. Sadda said.
“Regarding the dose and regimen of ranibizumab, there did not appear to be any effect of the dose on the development of atrophy. However, there did appear to be a trend toward a higher risk of development of atrophy in patients treated with mandated monthly treatments, regardless of the presence or absence of subretinal fluid,” Dr. Sadda stated.
He also pointed out that patients who received as-needed injections, who received more injections than the patients treated monthly, had less atrophy.
“There is more to this story that must be uncovered,” he said.
In addition to being a post-hoc analysis, other study limitations included the absence of a control group, no optimal detection methods, no consensus diagnostic criteria, and no quantitative progression data.
“It remains unknown if anti-VEGF drugs influence the development of atrophy in neovascular AMD,” Dr. Sadda explained. “VA gains do occur over time in patients treated with anti-VEGF drugs. Risk factors for development of atrophy have been identified. Right now, the treatment of these patients probably should not change.”
However, more questions remain.
Dr. Sadda pointed out that the plan for the future includes development of consensus/optimal methods for atrophy identification, re-analysis using a multimodal approach including optical coherence tomography, longitudinal quantitative analyses based on growth and not just the presence of atrophy, and incorporation of optimal methods to identify and track atrophy in future prospective studies.