Q&A: Deepak Sambhara, MD, discusses real-world follow-up for patients with GA

Photo of Deepak Sambhara, MD, taken at Retina World Congress 2025

In a comprehensive study of geographic atrophy (GA) treatment, Deepak Sambhara, MD, from the Eye Clinic of Wisconsin analyzed long-term patient compliance with two recently FDA-approved complement inhibitor therapies: pegcetacoplan (peg) and avacincaptad pegol (ACP).

Note: This conversation has been lightly edited for clarity.

Ophthalmology Times: Can you highlight the key points from your presentation here at Retina World Congress?

Deepak Sambhara, MD: In 2023 it was a pivotal year for geographic atrophy, because for the first time, we had two agents that were FDA approved to treat this disease. Now both drugs work by slowing the rate of GA lesion growth by binding and blocking the complement cascade at different nodes. But despite being on treatment, visual acuity and visual function continues to decline, which brings up the question of long term attrition, compliance and follow up for patients started on intravitreal treatment for geographic atrophy. That was the exact motivation for our data set.

This was a retrospective multicenter database review using the Vestrum Health database. We were looking at long term compliance and attrition of people who were started on complement inhibitor therapy with pegcetacoplan (or peg) or avacincaptad pegol (or ACP). We had a data cut off point through March of this year.

Interestingly, the baseline demographics of our population was about 20,000 eyes from 15,000 patients, and it was about a 2:1 split from peg to ACP, probably due to the fact that peg had an earlier market access availability due to earlier FDA approval. But if you look at baseline characteristics otherwise, there's about a third of eyes that had study eye CNV at baseline, almost 40% had fellow eye CNV at baseline and visual acuity was similar in about 55 letters on ETDRS, which translates to about 20/80. Over 50% of eyes had subfoveal involvement at baseline.

Because this is a longitudinal data set, it's really important to understand the spread of new starts to understand follow up opportunities that were available to each individual eye on the patient level. Now we define continuing treatment as exposure to drug up until a certain time point, whereas discontinuation of treatment was defined as greater than 120 days without drug exposure.

Now interestingly, you can look at trends seen on when new starts happen, and correlate them to real world events. For example, in July of 2023 there were fewer starts on peg, and that coincided with the time when the ASRS ReST committee released their position statement on peg. In 2025, January to present, we see fewer new starts across the board, and that could be correlated to the market access issues with third party nonprofit payers like the Good Days foundation.

Now the big question is, what happens with attrition in the real world? Our dataset would suggest that 25% attrition rate is what we see 6 months in irrespective of treatment, and that number swells up to as high as almost 45% through 18 months. So if we know that attrition is significant and it's real, the next natural question is, well, are there risk factors we can use to predict based on baseline characteristics, who's more likely to discontinue treatment? In order to answer that question, we ran a mixed models, Cox regression, hazards analysis, and what we found was there were 3 key features that predicted discontinuation.
Number 1 was study ICNV at baseline. Number 2 was fellow ICNV at baseline. And number 3 was very poor vision, 20 letters or less on ETDRS, which corresponds to about 2400 vision or worse. Now we tried to control for the fact that Good Days funding–or these third-party nonprofit issues–could have affected new starts. So we looked at the data from January to March of this year, and we tried to see if payer access issues affected new starts. And we didn't really find anything on that Cox analysis, but this is an ongoing analysis as we accrue more data over time.

So in conclusion, we know that significant attrition happens in folks receiving intravitreal, complement inhibitor therapy. We know that it's hard to make an apples-to-apples comparison across the 2 agents–ACP and peg–but we have to understand that this is something rampant, seen across both agents. And finally, we know certain risk factors predispose patients to earlier discontinuation, and it's really imperative for us to longitudinally follow these patients and this dataset over time to understand the motivating factors for discontinuation. Is it physician or patient fatigue? Is it adverse events? Is it payer access or market issues? And we can better understand why these trends are happening that we see in this data set.

OT: When we look at real-world studies like this one, what value can this type of data bring to our understanding of the condition and its current treatment options?

Sambhara: In some respects, large database studies can make informed decisions for how we understand trends that we may not be realizing on the individual practice level, but if we see these global trends start to emerge, it helps us identify where gaps exist and how we can potentially bridge those gaps. And so for every problem that exists, it's actually an opportunity that's created. In the world of geographic atrophy, it's difficult because we don't have instant gratification. The best revision is ever going to be is today, because it's a continually progressive disease.

In order to create a culture of compliance where patients are locked into their treatment, we really need to rethink the game, because this is different than the anti-VEGF agents that we're so used to delivering on a day to day basis, where patients can see that "aha" effect of a before and after on a picture, or notice their own vision improving after treatment. So it gives us an opportunity to identify unique ways to create a culture of compliance in an otherwise population that's getting worse, that doesn't necessarily have the same motivating factors to keep continuing treatment.