In Complement Inhibition Debate, Game Change Goes Against Too Little, Too Late | AAO 2025

Ophthalmology, like all medical specialties, has its share of disagreements and controversies — some heated, many knotty and technical — about diagnosis, treatment and a host of other issues.

A debate about one of the most contentious was held at the American Academy of Ophthalmology’s annual meeting yesterday in Orlando, although the supposed opponents seemed to have large areas of disagreement.

Neil Bressler, M.D., a professor of ophthalmology at Johns Hopkins and editor-in-chief of JAMA Ophthalmology, argued that complement inhibition for geographic atrophy is a game changer. His debating partner, Stephen J. Kim, retina division chief at Vanderbilt University Medical Center, argued for complement inhibition, "too little, too late." After the debate, approximately 80% of the audience indicated that they agreed with Kim's too little, too late position.

The complement system is part of the immune response and also plays a role in inflammation. In 2023, the FDA approved two drugs, Syfovre (pegcetacoplan) and Izervay (avacincaptad pegol), as treatments for geographic atrophy, a form of advanced-stage dry age-related macular degeneration. Both drugs are believed to work by inhibiting the complement system and, in so doing, inflammatory processes that play a role in the development of geographic atrophy.

On the one hand, Syfovre and Izervay were heralded as major advances as FDA-approved treatments for a condition for which there were none. On the other hand, both were approved because they slowed the growth of the geographic atrophy lesion. Neither has shown benefit as far as vision is concerned. Much of the controversy about them is centered on the discrepancy between the anatomic benefits that are apparently unattended by functional ones.

Bressler didn’t shy away from acknowledging the lack of vision benefit. The FDA approval of pegcetacoplan — he used the generic name and focused on pegcetacoplan — was based on it reducing the growth of geographic atrophy by 20% relative to the growth among those in the placebo through two years. “But no prespecified functional vision benefits were noted compared with sham,” he said, noting that the lack of difference was not limited to best corrected visual acuity and extended to secondary outcomes such as maximum reading speed and macular sensitivity by microperimetry. Bressler noted the risks, including choroidal neovascularization and rare cases of occlusive retinal vasculitis. He noted that pegcetacoplan has not been approved by European drug regulators (neither has Izervay).

But Bressler stuck to his assertion that complement inhibition is a game changer. “Why?” he asked rhetorically. “Because geographic atrophy is not a game.” Bressler said that approval of the complement inhibitors had brought “scientific attention” to a condition that results in thousands of people going blind each year. He noted that at the academy’s retina subspecialty day in 2014, there were no clinical trial results reported and only one talk about complement inhibition. This year he said there were six talks about geographic atrophy trial results and a panel to discuss geographic atrophy in clinical settings.

Wrong target

Kim said he felt like he had a stronger hand and thanked Bressler for “supporting his talk.”

Kim echoed Bressler’s mention of the evidence showing anatomic gains without visual ones. But he also cast a wider net of doubt on the complement system inhibitors, including whether complement inhibition was a flawed strategy in the first place. “There’s also no direct evidence that the complement has an involvement in geographic atrophy, so why would we think this one system would be a silver bullet?” Kim said.

Kim had made several arguments supporting this “too little, too late” position on complement inhibition. Noting that geographic atrophy progresses slowly and tends to spare the fovea till late in the disease process, Kim argued that simply observing the condition might be preferable in some cases. The average age of the participants in the two pivotal trials of pegcetacoplan was 78, he said. Based on life expectancy tables and lack of visual benefit, many patients may not have any “meaningful clinical benefit with treatment” in their remaining lifespan, Kim said.

Kim also said there was reason to question the data from the pegcetacoplan trials because of the relatively high rate at which participants left the trials or were lost to follow-up and the resulting attrition bias. OAKS and DERBY were the large phase 3 trials of pegcetacoplan, and GALE was the open-label study of the drug.

“Let's do the math,” Kim said. “If 70% of people completed OAKS and DERBY, 83% of them were recruited [for] GALE, and 90% of those would finish one year, we only have 53% compliance overall,” he said.

Kim also called attention to the possible harms, including the rate of macular neovascularization. “And equally concerning, in my opinion, is we don't know why,” he said. “Why is this happening? This runs counter to what we would think the mechanism [of] complement [inhibition] would be in these conditions.”

Kim said ophthalmologists have an obligation to provide treatments with “a clear, measurable benefit that exceeds the risk, burden and cost of treatment.” For him, he said, there are a lot of question marks about complement inhibition, “and because of that, I'm staying on the sideline for now.”