Geographic Atrophy

Complement Inhibitors and AI for Geographic Atrophy: An Expert Weighs In

Posted on

Complement inhibitors are emerging as a potential treatment option for geographic atrophy (GA). Paired with artificial intelligence (AI), these medicines will play an increasing role in managing individuals with GA. We asked a vitreoretinal expert Mitul Mehta, MD, MS, FASRS, Clinical Associate Professor of Ophthalmology at the Gavin Herbert Eye Institute, University of California, Irvine for his thoughts on using genotypic features for treatment decisions, the role of AI, current challenges and best practices, and the road ahead.

Do you think there’s a place or will be a place for clinicians to use genotypic features to make treatment decisions for GA?

Dr Mehta: I use them all the time, though currently, I am in the minority. I lead the America Retina Forum, which conducted a poll where 60% of members said they find no value in treating GA. That’s surprising. These medications can help certain patients, and using genotypic features is a way to identify them.

How do you determine who is the ideal candidate for treatment?

Dr Mehta: I look for GA that is growing or close to the fovea. I exclude individuals with autoimmune diseases or known retinal vasculitis because a certain percentage can develop an autoimmune reaction, leading to severe vision loss. I also treat those who lost vision in one eye. I start by testing the bad eye. If there are no side effects after a few weeks I will treat the other eye.

Do you think AI will help in the management of individuals with GA?

Dr Mehta: One hundred percent. In fact, it is already happening in ophthalmology. A study published in April 2024 showed that GPT-4 held its own in performing vision assessments compared with ophthalmologist assessments.

AI will be able to sort through high-density retina scans very quickly to identify ideal treatment candidates. It will speed things up versus having a human scroll through images and make determinations. Busy clinics have bandwidth issues that AI is going to help alleviate.

What other challenges do you face managing GA?

Dr Mehta: Heterogeneity is a challenge, demographically speaking. Most of the US studies involve participants of European descent. But I treat a lot of individuals of Southeast Asian descent; they respond differently to these treatments. Complicating matters is that America is a melting pot where people have mixed backgrounds. That is one of the challenges with AI, which bases its recommendations on study results. But our patients are unique, so the human element remains important.

Where do you see the management of GA heading in the next year?

Dr Mehta: I don’t think it will change much. Technology and drug development tend to move slowly. As we get more data, more retinal specialists will become comfortable with treating GA. But don’t forget that low vision technology is available to help people with vision loss from GA see better. Augmented reality glasses, closed circuit televisions, handheld magnifiers, head-mounted telescopes, and virtual reality solutions can be used even in patients who are treated medically.

What about in the next five years?
Dr Mehta: In the next 5 years more drugs will become available. We may have a drug that inhibits C1q (ANX007 is currently being evaluated). There could also be a role for photobiomodulation. It has been evaluated in individuals  with age-related macular degeneration. We may see a trial involving those with GA.

This content is independent editorial sponsored by Astellas. Astellas had no input in the development of this content.