Baseline ellipsoid zone integrity features can predict rate of geographic atrophy progression

Katherine Talcott, MD, retina specialist at Cleveland Clinic at the Cole Eye Institute, talked about the results of a study presented at the ASRS 2024 Annual Meeting that assessed the importance of predicting disease progression in patients with geographic atrophy (GA).

Question:

You recently presented a study at the American Society of Retina Specialists (ASRS) Annual Meeting titled, “Baseline Ellipsoid Zone Integrity Features as Predictors of Geographic Atrophy Growth Rate in the Phase 3 GATHER Clinical Trials.” Can you describe the study’s goals and design?

Katherine E. Talcott, MD, FASRS:

I think this is a really important study that we were able to undertake, because we’re so excited to be in an era now where for geographic atrophy, we now have treatment. It’s a little difficult for us as clinicians to be able to tell in general, for people with advanced AMD and geographic atrophy, who’s going to progress and who isn’t, and then as a corollary, who might benefit from these therapeutic interventions and who might have less benefit, so we can tailor expectations as well as select people for treatment. This study was a post-hoc analysis of the GATHER 1 and GATHER 2 studies to be able to better understand which of the patients are going to get worse over a quicker period of time.

Our goals for this study were really to look at patients who were in the GATHER 1 and GATHER 2 studies, and to be able to better understand what’s going on in the retina outside those areas of geographic atrophy. One of the things that you can look at is the ellipsoid zone (EZ), and we wanted to be able to look at patients and understand better who of those patients had more EZ loss and who had less EZ loss, and how that correlated with the growth rate of their geographic atrophy.

Question:

This study used patient cohorts from the phase 3 GATHER trials. Can you briefly describe those studies?

Katherine E. Talcott, MD, FASRS:

The GATHER 1 and GATHER 2 studies are pivotal clinical trials looking at avacincaptad pegol to be able to tell if it was able to slow growth rate in geographic atrophy. Utilizing patients as part of a post-hoc analysis for this type of study is a really nice opportunity, because these patients with extrafoveal geographic atrophy had a lot of imaging that was done and was followed in a very systematic way. It’s a really nice way to be able to tell, even from the therapeutic effect of the drug, and better understand the natural history of certain patients. That’s what our primary goal was, but then our secondary objective was also to be able to see if avacincaptad pegol actually was able to slow the growth rate of geographic atrophy in those patients who had more EZ loss at baseline.

Question:

What were the findings of this recent study presented at ASRS?

Katherine E. Talcott, MD, FASRS:

In the study, we looked at a couple different outer retinal metrics. We looked at EZ attenuation in general within the macula. We also looked at what’s called an EZ-GA gap, which is basically looking outside of the areas of geographic atrophy to be able to see if there was EZ attenuation in those areas as well.

What we did is we looked at patients in the study and we stratified them by the growth rate of their GA. As we know from taking care of these patients clinically, some people don’t have any change in GA, and a lot of patients have a lot of change in GA. We stratified them into GA growth rate by quartile, and then we looked within those quartiles to be able to see what kind of EZ attenuation they had at baseline.

We found that, for each of the different EZ metrics, patients who had the greatest growth rate of their geographic atrophy also had the most attenuation of their EZ at baseline. There seemed to be a correlation there, and there was a significant difference between the patients who had the most growth rate of their GA to the patients who had the least growth rate.

We found that that was to be the case for most patients who were treated with sham, as well as the patients who were treated with avacincaptad pegol (ACP). We then looked at the patients who had more EZ loss at baseline, and we found for those patients that there was a significant difference between the patients who were treated with sham in terms of their GA growth rate and the patients who were treated with ACP.

Question:

How might the rate of geographic atrophy progression inform clinicians when it comes to planning treatment and speaking to patients?

Katherine E. Talcott, MD, FASRS:

If we’re thinking about it, patients will often want to know, as they come in with geographic atrophy or AMD, how quickly this is going to change. Are they going to be able to maintain good vision for a long time, or is it something where they’re going to have more trouble functioning; they might not be able to drive for as long or take care of themselves in the same way that they were. Patients really want to know what does the future look like for them.

Similarly for us as clinicians, now that we have treatment options that are available that can slow the growth of their GA, we want to be able to better identify patients who would benefit from those therapies that would require patients to come in every month or two into clinic. Having a better understanding of who’s going to progress can help us target patients that we might want to watch closer or potentially start treatment on. Studies like this one to be able to better identify patients who are at higher risk are important and helpful to be able to do.

Question:

How might these findings impact geographic atrophy care?

Katherine E. Talcott, MD, FASRS:

We need to be able to better put together models to be able to identify who’s going to progress faster. This study looked at new metrics that can be able to identify these patients, but really the next step is coming up with some sort of automated, machine learning-type model that we would be able to more easily put into clinic to identify patients at risk for growth.

That’s the next step. It’s finding out a way that we can make information such as we found in the study more clinically relevant, some way to be able to better utilize the metrics, looking at EZ attenuation that we looked at in the study, but being able to come up with a system in which, if you’re seeing a patient in your clinic, a system could look at the images and in real time be able to tell you who’s at risk for progressing and who’s not. That’s sort of the next step to do.