Can deep phenotyping using retinal images predict response to intravitreal aflibercept therapy in patients with neovascular age-related macular degeneration?

Wet age-related macular degeneration (wet AMD) happens when abnormal blood vessels grow and leak fluid or blood into the retina at the back of the eye. Without treatment, this can quickly damage your central vision, making activities like reading or recognising faces difficult. In the NHS, the standard first-line treatment is a medicine called aflibercept. It is given as an injection directly into the eye, starting with three doses spaced one month apart.

Every person's eye responds slightly differently to this treatment. For about four to five out of ten patients, the medicine works very quickly and clears all the fluid from the retina. Once the eye is completely dry, their vision is well-protected. However, doctors currently have no way of knowing in advance who will have this excellent response and who might need closer monitoring or a different approach.

This research study is investigating whether modern technology can solve this problem. When you attend an eye clinic, staff take detailed 3D pictures of the back of your eye using painless scanners called OCT and OCTA. Researchers at Moorfields Eye Hospital want to see if artificial intelligence (AI) can analyse these routine images and spot subtle patterns that predict how well the eye will clear up.

By comparing the AI's findings against assessments made by specialist eye doctors, the team hopes to prove whether AI can accurately forecast treatment outcomes. If successful, this could help doctors provide patients with clearer expectations from day one and pave the way for more personalised eye care in the future.