Mark Barakat, MD

Changing Dosage Regimens in Patients with Diabetic Macular Edema, with Mark Barakat, MD

2025 Association for Research in Vision and Ophthalmology (ARVO) Annual Meeting

, Mark Barakat, MD, founder and director of research at Retina Macula Institute of Arizona, presented a post-hoc analysis examining the shortening or extending of aflibercept 8 mg dosage through week 96 of the PHOTON trial.

“So here we have an agent that’s been shown to go out to not just 12, 16 weeks, but in some patients even longer, like 20 or 24 weeks,” Barakat told 

. “Right now, we’re very fortunate to have anti-VEGF agents, which really give us good results. But durability is king. So here we have something that can stretch things out longer. From a patient’s perspective, it’s a lot less injections, perhaps fewer visits.”

 were randomly assigned to aflibercept 8 mg every 12 or 16 weeks (8q12 or 8q16, respectively). After 3 monthly doses, patients were eligible for dosing modification so long as they met prespecified criteria. Dose interval shortening to a minimum 8 weeks was permitted in years 1 and 2; dose interval extension to a maximum of 24 weeks was only permitted in year 2.

Barakat and colleagues noted that dosing intervals were shortened for 12.5% and ≤16.5% of patients treated with 8q12 (n = 256) and 8q16 (n = 139) who continued until Week 96. The shortened 8q12 group exhibited best-corrected visual acuity (BCVA) at baseline and Week 96 of 61.5 and 66.9; the maintained group exhibited a 63.5 and 72 BCVA, and those extended exhibited 64.4 and 74.4 letters.

Additionally, investigators indicated that corresponding retinal thickness (CRT) was 509.1 and 333.9, 488.2 and 330.2, and 431.1 and 228.3 µm, respectively. BCVA at baseline and week 96 for the shortened, maintained, and extended subgroups was 55.4 and 62.5, 62.7 and 68.3, and 63 and 72.2 letters, respectively. Corresponding CRT was reduced from 521.5 to 341.5, 472.2 and 343.3, and 418.6 and 256 µm.

A combined 8 mg analysis revealed that baseline factors with interval shortening versus maintained or extended were shorter duration of diabetes (OR [odds ratio] .85; 95% CI, .73-1 for every 5-year increase) and thicker CRT (OR, 1.18; 95% CI, 1.07-1.31] for every 50-µm increase). Lower baseline CRT was associated with interval extension versus maintained (OR, .82; 95% CI, .74-.91 for every 50-µm increase).

Ultimately, Barakat and colleagues indicated that all dosing intervals exhibited substantial BCVA gains and CRT improvements at week 96. Patients with shorter diabetes durations or thicker CRT at baseline were more likely to have dosing intervals shortened; however, those with thinner CRT at baseline were more likely to have their dosing intervals extended.

“Frankly, apart from who gets shortened and who gets extended and maintained, the important thing is, if you look back at it, all these patients did really well in terms of anatomy and vision,” Barakat said.

Barakat reports the following disclosures: Oxurion, Janssen, RegenxBio, Novartis, Biocryst, and others.

Barakat, Mark R. Baseline characteristics and outcomes of patients treated with aflibercept 8 mg at shortened, maintained, or extended dosing intervals through 96 weeks in PHOTON. Abstract presented at Association for Research in Vision and Ophthalmology in Salt Lake City, UT, from May 3 – May 8, 2025.

Videos

A post-hoc analysis has indicated that aflibercept 8 mg has the potential to shorten or lengthen intervals between treatment in patients with DME.

And so before we go further, what's shared decision-making? It's basically the health provider and the patient have to work together to come up with a decision that's not best for the clinic, it's not best for the provider, it's best for the patient. And it's really all about knowledge. The knowledge base that we have is on the medical side. The knowledge base that the patient has is on the cultural, on the socioeconomic aspects, on values. All those things have to come together. We're going to talk about this a little bit more detail later on, but this is the SHARE approach. There are 5 steps: seek your patient's participation, help your patient explore and compare treatment options, assess your values for the patient, the preferences, then reach a decision. But don't forget step five. Every now and then you got to step back and evaluate the decision and see where we stand. And this is based on these 9 essential elements, you have to look at the problem, you have to present options, risk benefits, costs, values of the patient's preferences, the ability, self-efficacy of the patient, the recommendations, of course, the medical recommendations, right? Are you sure the patient's understanding? And then you make a decision and then you arrange for follow-up and re-evaluation.

But, again, why? We're here to talk about the data, right? Why do we care? We're busy. We're seeing a million patients. Why do we care? Because it helps everybody. It improves the patient's experience. It makes them more likely to adhere to treatment, to follow-up, to not get lost to follow-up. It increases the quality of care delivered and it increases the patient satisfaction. Well, great. That's wonderful for the patient. That in and of itself should be reason enough to do it. But frankly, it's good for the clinic, too. And I found this really interesting, because I am probably guilty of this. Many healthcare providers believe patients are not interested in participating in the decision-making process. There's actually this literature that suggests that patients actually want more information than they're typically given. Many of them would want to be more involved with the caveat that we all have that patient that does not want to, or is not ready to be part of the decision-making process. And, you know, that's something that we're respected. You will get that patient that says, “Doc, whatever you think is best for me, go ahead and do that.” That in and of itself is a decision, too. But on average, many people want more information than we think.

Well, how do we do this? Well, we were talking about the SHARE approach. Step 1, seek your patient's participation. That's probably the easiest thing to do. “Hey, do you want to be in a dialogue?” “Do you want to be involved in making this decision?” Or, do you fall into that category of, “Hey, doc, do what's best for me?” Start the dialogue. Step 2 is risk-benefits alternatives. That's an important consent in my mind. Discuss the benefits and harms of each option. We do this all the time. Sometimes a little bit more thoroughly than others. Possibly depending on how many patients you're seeing that day. But risk -benefits alternatives. Step 3, this is where it gets interesting. This is where we may not have the time or the training to do as well as we otherwise might. Assess your patient's values and preferences, and if they are taking into account what matters most to them. Now, I think many times, right now, specials do a great job with that. For example, if you have a patient with wet macular generation, you have this form of scar. And I'm willing to bet you have patients that say, “I don't care if I have this form of scar. I understand that it's not going to get any better. But I'm worried that it might get a little bit bigger. Keep going.” Or, if other patients say, “I don't want to keep going. That's not my decision.” That's their values, right? And we have that discussion all the time.

Then step 4, you reach a decision. And step 5, every now and again: “Okay, Ms. Jones, you've decided to keep injecting that this form of scar. We've been doing it for X number of months. Are you still happy with that? Do you see a difference?” You got to reevaluate it now and then it's not carved in stone. Patient buy-in is really essential to adherence. Now, without belaboring this anymore, there's so much more. This, this in and of itself could be like a 10-hour talk. But we do want to touch on just a few hurdles, because it seems so easy when you think about it. But there's more than a few hurdles. One, language, hearing, eyesight; [these are] common barriers. We all know about the vision. That's why we're here. But 9 out of a 100 limited English proficiency; 2 out of a hundred visual disability, 3% hearing disability, qualified medical interpreters. We have a bevy of qualified medical interpreters, which are listed here. Now, having a family members translate is probably not the best idea. And so that is one thing that we need to think about: health literacy. Twelve percent of US adults age 16 and older have proficient reading skills. Thirty-six percent of the population lack adequate literacy skills. And [that’s] for all ages, races, and income. So how can you get informed consent? It's really a conundrum that needs to be addressed. The impact is disproportionate. It affects lower socioeconomic and minority groups more so than it does others, so you have to keep that in mind. And here's the kicker: Nearly 9 out of 10 adults lack full literacy and have difficulty using everyday health information. It's shifting gears and trying to understand where the patients [are] coming from. Because here's the reason why: they are more likely to skip tests, underuse healthcare, to visit the emergency room, be hospitalized, not comply with treatment plan, etc.

The issue is patient buy-in. Here's some of the strategies. Unlike me, you should probably speak slowly and not rush through everything. Show some empathy. Avoid medical jargon. I think I'm as guilty of that as the next person, right? You're using plain language. The teach-back technique, I think is wonderful. It just takes a lot of time asking, “Hey, Mrs. Jones, what did I just tell you? What did we agree on?” That's hard, but it may be worth it, putting that effort in that first time and then read the benefits.

I think we're all really good at thinking about literacy, but what is numeracy? The ability to use numeric information in the context of health; it is also limited. Have you ever had a discussion with a patient and say, “Well, there's a 1 in 4,000 risk of [a certain side effect] or there's a 0.5% risk of [another side effect]?” And do you ever get that look [when you know you’ve] lost them. That's exactly what it is, because you need to understand numbers when choosing treatments, when weighing the risks and the benefits, right? If Matt and I were all happy, we’re thrilled, if close to 80% of patients can go up to Q12 to Q16 weeks. That’s a lot of numbers coming at you. You need these numbers for a bevy of tasks that are shown here. Here are some of the strategies. Frankly, use them only when you need them. When you need precision, use everyday words and approximations; [for example,] say “about one half” instead of 48%. You can elaborate for providing estimate numbers. Visual aids are great. Absolute risk instead of relative risk. That's a beautiful one, right? If you say, well, there's an additional small risk of X% and I do this all the time, just say, well, it's a tiny percent additional risk.

Transcript is AI-generated and edited for clarity and readability.

Mark Barakat, MD, reviews the best practices for patient communication and shared decision making.

Patient Communication and Shared Decision Making

 This is about TENAYA and LUCERNE, which looked at faricimab dosing up to 16 weeks compared to aflibercept with a regular loading dose and then Q8 week dosing. Essentially, there were OCT and visual acuity criteria for determining if the patients could go Q8 weeks, Q12 weeks, or Q16 weeks on the faricimab dosing. So akin to, if you want to say, a more rigorous treat-and-extend that we do in the clinic. You want to make sure that the anatomy matches the vision, and the vision matches the anatomy. And you'll see this, again, this nice response in both the faricimab arm and blue and the aflibercept arm, and we're very lucky that aflibercept is a great drug. We can see it here, as well. And you see that they have a nice response all the way through week 48, but with fewer doses of the faricimab. This is just a percentage of patients with absence. So basically, are they dry? It's not just a question of the CFD, which is a number that I may not even look at in the clinic. I may look at just the anatomy. And if you look, weeks 4, 8, 12, and 48, it's actually in favor of the faricimab arm in terms of it being dry in terms of absence, somewhere between 60% and 77%, and with the aflibercept arm you get somewhere between 47% and 67%.

These are the key results. We talked about on the left-hand side, we talked about the vision. In the middle, we talked about anatomy. On the right-hand side, we talked about something that we care very deeply about as clinicians. How many of these patients can stay about 12 to 16 weeks? About 80 percent. This is the anatomy. As you can see here, nice stability throughout in both TENAYA and LUCERNE, all the way through week 112. Now, interestingly, if you look at year two, right, this is not, again, being achieved with 6 injections of aflibercept and 3 injections of VABYSMO. So there's a difference. Comparable proportions of patients gained or avoided loss. So, you know, the same thing we talked about before, different ways of looking at it. The majority of patients in TENAYA are roughly 90% in both the aflibercept arm and the faricimab arm maintain vision, meaning didn't lose three lines. Same thing with LUCERNE, and it's about 20% or so in both trials, in both arms, over three line gainers. So, again, a nice response. Here it is. Greater than 60% of faricimab patients achieved Q16 weak dosing at the snapshot, at week 112. So, basically, if you continue with these patients to be flexible in terms of the treatment regimen, as you can see, 74%, 81% can go to Q12 to Q16. So, this here is going to be the vision gains baseline with Vabysmo. Up to 16 weeks, they were comparable, and this is pooled. Just to keep it equal, because I showed you the year one in pooled. This is very encouraging.

Now, if you want to get a little bit more granular, you ask yourself the heat map on the left-hand side. So, if you look at the greenest initiation period, and then if you look at the purple is Q12, and the pink is Q16, and you see the line, the dividing line in the middle of year one. So, we know the ones that are highlighted in the green box right there are the patients that at year one, they got to 16 weeks. weeks. End of story, right? Because it's a one-year disease, we know that. The question is, how many of them actually stayed at 16 weeks? the majority, about 70%. And if you look at the anatomy, if you look at the vision to the right, you'll see that they stayed, they were nice and stable. Well, great. Well, how about those patients that weren't lucky enough or well-treated enough at year one? They were not at 16 weeks, but they were 12 weeks, because life's a bell curve. How did they do? Well, if you started at 12 weeks at year 1, at year 2, again, the majority were either at 12 or at 16. Some of them stretched out, we look at 60%. Again, nice anatomy. And you can extrapolate that to those folks that need a more frequent injections. Of course, I wish I could say that for SNAP, everyone is on Q12, we know better than that. There are some that need Q8, but thankfully, for people that needed Q8 treatment at year one, the majority of them were actually on increased dosing interval by the end of year two. So, you know, to Diana's point, we have to treat, we have to treat rigorously, but that doesn't necessarily mean that we always have to maintain the same interval because we have data here that shows that you can extend and safely, as you can see on the right-hand side.

Speaking of safety, safety is foremost in all of our minds. We do no harm. This was very well tolerated. If you look at the safety results in both the faricimab arm on the left, the aflibercept arm on the right, it was well balanced. And we always worry about inflammation. There really was no inflammatory signal of note. Of course, there were cases of endophthalmitis.

In summary, we have good data now about individualized dosing up to 16 weeks in control of anatomy, in control of vision, in patients who years, two years of follow-up in macular generation and the important thing is this is not new we know we can do this we know we can do this is a monthly injection so we know we can do this is a bi-monthly injection but this was done with about 80% of solo patients at Q12 week injection and 60% at Q16 injection, so well tolerated, and as we'll hear later from Matt we'll talk about the long-term extension study and other studies as well so there is a commitment to actually further follow-up to help us understand these new molecules better.

Mark Barakat, MD, discusses the phase 3 TENAYA and LUCERNE trials for patients with neovascular age-related macular degeneration (nAMD).

Mark Barakat, MD

Articles