I'm Vas Sadda. I'm professor of ophthalmology at Doheny Eye Institute and UCLA and I've had the pleasure of attending ARVO 2026 in Denver. I had an opportunity to present on an interesting post-hoc analysis that we performed on data from the TENAYA and LUCERNE studies of faricimab in particular the talk focused on subretinal hyperreflective material. This has actually been a topic of great interest, personal interest, to me because we've first studied this almost two decades ago now, it seems, at least 15 years ago and back in the time domain OCT era, when we identified that this material in the subretinal space, we called it subretinal tissue in those days. But in any event, we were the first to show that that was one of the most important predictors of visual outcomes in patients with neovascular AMD. And it kind of makes sense because you could imagine that this material between the photoreceptors and the RPE is probably not a good thing for photoreceptors. And so I think it's quite expected that it translates to poor visual outcomes.

So we had an opportunity to study this again in the context of these clinical trials. And so what we did was we performed this post-hoc analysis using a deep-learning algorithm to automatically quantify the subretinal hyperreflective material. So we could quantify its area, its thickness, its volume. And then we looked to see what could we actually predict by knowing that. And one of the things we confirmed was that again, patients who have persistent subretinal hyperreflective material, and you can look at different cutoffs having volume of less than 10 nanoliters or volume of less than five nanoliters. It really doesn't matter. The bottom line is if you have more of it, you tend to have worse visual outcomes. And when I'm talking about more of it, I'm talking about after the loading phase, for example, if there's some persistent SHREM or subretinal hyperreflective material, that translates to worse outcome for patients.

In addition, we also found that patients who have persistent SHRM after the loading phase, those are also patients who are less likely to be able to achieve a maximum extension interval. So they were les likely to get to 16 weeks, which was the maximum interval in the trial. So again, it's also a sign of worse disease in some ways. We also showed that patients who had type two lesions were more likely to have SHRM. That I think is common sense because type-two neovascular lesions are in the subretinal space.

And finally, and perhaps quite importantly for clinicians thinking about how we manage patients, we did find a difference between aflibercept and faricimab in terms of SHRM reduction with a greater reduction in both area as well as volume of SHRM associated with faricimab compared to aflibercept two milligrams. And this was especially the case in patients with type-two lesions.

And so this kind of information I think can help clinicians as they think, as they contemplate what types of therapies they may consider for patients.