Skip to content

Retina Experts Serve Up a Combination Platter at OIS@ASRS

Contributed By: Michael Lachman

The OIS@ASRS panel discussion on Emerging Approaches to Combination Therapies in AMD & DME was moderated by Mark Humayan, MD, PhD, director of the USC Eye Institute, and Emmett T. Cunningham, Jr., MD, PhD, MPH, of Clarus Ventures.

The need for new treatment approaches was articulated by Pravin Dugel, MD, Arizona-based retina specialist and clinical professor at the USC Keck School of Medicine. “From both a clinical and industry point of view, we’ve realized that, as good as VEGF-A suppression is, we’ve reached a ceiling effect and we’re starving for combination therapies. I’m excited about all of the approaches … we’re on the precipice of changing our treatment paradigm.”

New disease mechanisms under investigation were discussed by Donald J. D’Amico, MD, professor and chairman of ophthalmology at Weill Cornell Medical College. “The Tie2/Ang story is the one that strikes me as the closest to making sense, having data, having a mechanism that really looks like there’s a drug in there somewhere.” He continued, “The approach that I really still think we have short-changed is steroids. We have approved steroid preparations and implants. But all of the panelists would say that if we could recall and bottle, in a safe way, that wow effect that we got in some of these diseases with the initial intravitreal triamcinolone work, we would really want that in our armamentarium. And yet, it’s too much like a baseball bat. I would hope that we could continue our work of understanding precisely how steroids produce that effect.”

Age-related macular degeneration (AMD), diabetic macular edema (DME), and retinal vein occlusion (RVO) are very different diseases, noted Peter Kaiser, MD, retina specialist at the Cleveland Clinic’s Cole Eye Institute. Yet they’ve been treated in much the same way, because anti-VEGF addresses vascular leakage in all of them. “What works in AMD may not be the same thing that works in DME. For instance, the Tie2 pathway is very important in diabetes, probably a little less important in AMD, but that doesn’t mean a drug shouldn’t go forward.” Dr. Kaiser believes that in the future, we’ll more likely be talking about specific treatments for the different diseases.

Anti-VEGF-A drugs have been successful in treating numerous retinal diseases by addressing vessel permeability. The panel discussed the relative importance of addressing additional mechanisms, such as anti-angiogenesis, fibrosis, and neuroprotection. Dr. Kaiser noted that none of the current drugs for AMD is anti-angiogenic or reverses choroidal neovascularization (CNV), which may be why lesions become fibrotic. Dr. D’Amico pointed out that neuroprotection has been incredibly hard to pin down, and that addressing fibrosis is very important: “If you could preserve a retina in a reversible state, and not have it become a scar, I think you’d live to fight another day, so I’d put [anti-fibrosis] at the top of the list.” Dr Dugel said that we have a “biomarker crisis” when it comes to addressing new disease mechanisms. “That’s the biggest challenge that we have in drug development, that we simply don’t have a good biomarker for these things as we did for permeability.”

Mark Blumenkranz, MD, professor of ophthalmology at Stanford University and serial entrepreneur in medical technology, commented that in addition to combination drug therapy, clinicians could be doing a better job with multi-modality therapy, such as combining intravitreal drug injections with modalities such as lasers, surgery, and gene therapy. “I think that mixing and matching these things works well.” Dr. Blumenkranz also said that differences in drug dosing duration between one month and six months “are primarily tactical advantages that you have over the short term. I think what people really are looking for is to cure these diseases, and probably the best way to cure a disease is to prevent it. It’s much harder to cause regression of an established disease than it is to prevent it. The catch is that you have to know who’s going to get it, then you have to have a therapy that’s safe, effective, and relatively noninvasive. It’s within our reach to do that right now.”

Other panelists concurred with Dr. Blumenkranz’s view that efficacy is more important than dosing frequency. “Vision trumps all,” said Dr. Kaiser, “whatever provides the best vision, that’s the drug we’ll use.” Added Dr. Dugel, “At the end of the day, efficacy is king, efficacy always wins. I’m old enough to remember when we said, ‘Who would ever take a needle in their eye once every four or five months?’ and look what we’re doing now.” Dr. D’Amico also expressed a preference for being able to use these new drugs individually, rather than being forced to combine them from the start.

The Wall Street perspective regarding the types of companies that can be funded today was provided by Adnan Butt, biotech analyst with RBC Capital Markets. At this time, companies that improve clinical outcomes as well as those that improve duration of effect – with six months being ideal – are fundable. New mechanisms of action are of great interest because they have the potential to move away from Avastin as the retina drug of choice.

Dr. Dugel concluded the discussion of combination therapy on an optimistic note. “I look forward to having a portfolio where we’ll be able to truly individualize and customize care. I think that’s the ultimate goal.”



Emmett T. Cunningham Jr., MD, PhD, MPH

Dr. Cunningham joined Clarus Ventures with 20+ years in the biomedical and biopharmaceutical sectors. Prior, Dr. Cunningham was the SVP, Medical Strategy at Eyetech Pharmaceuticals.

View Full Profile


Mark Humayun, MD, PhD

Mark S. Humayun, MD, PhD, is the Cornelius J. Pings Chair in Biomedical Sciences, Professor of Ophthalmology, Biomedical Engineering and Cell and Neurobiology, Director of the USC Institute for Biomedical Therapeutics, and Co-Director of the USC Eye Institute.

View Full Profile

Mark Blumenkranz

Mark S. Blumenkranz, MD

Mark S. Blumenkranz, M.D., MMS, is the HJ Smead Professor in the Department of Ophthalmology at Stanford University. He served as the Department Chairman from 1997 until 2015 and played a leading role in the construction of the Byers Eye Institute.

View Full Profile


Adnan Butt

Adnan is a senior biotechnology analyst at RBC Capital Markets. His current coverage universe and research interests encompass cardiology, infectious diseases, neuro-psychiatry, oncology, and ophthalmology, as well as other emerging areas and modalities.

View Full Profile


Donald J. D’Amico, MD

Donald J. D’Amico, MD, Professor and Chairman of Ophthalmology at Weill Cornell Medical College and Ophthalmologist-in-Chief at New York Presbyterian Hospital, is an internationally recognized leader in vitreoretinal diseases.

View Full Profile


Pravin Dugel, MD

Dr. Dugel is a Clinical Professor at Roski Eye Institute, Keck School of Medicine at the University of Southern California (USC), Managing Partner of Retinal Consultants of Arizona, Founding Member of Spectra Eye Institute and Physician Executive Director of the Banner Eye Institute, Banner University Medical Center.

View Full Profile

Peter Kaiser

Peter Kaiser, MD

Graduated magna cum laude from Harvard College and Harvard Medical School. He completed an internship at Massachusetts General Hospital, an ophthalmology residency at the Massachusetts Eye and Ear Infirmary, and a vitreoretinal fellowship at Bascom Palmer Eye Institute before joining the Cole Eye Institute where he is the Chaney Family Endowed Professor of Ophthalmology Research.

View Full Profile


Emmett Cunningham: As we get ready, perhaps each of you can just state where you are and your affiliation and title. I think we know all of you, but just for the few who might not have met you.

Mark Blumenkranz: I’m Mark Blumenkranz. I’m a Professor of Ophthalmology at Stanford. I’m also a serial entrepreneur in the biomedical device and drug space.

Don D’Amico: My name is Don D’Amico. I’m Professor and Chair at Weill Cornell New York Presbyterian. And I’m up here to give Mark all the answers.

Pravin Dugel: I’m Pravin Dugel. I’m a retinal surgeon in Phoenix, Arizona and a clinical professor at the USC School of Medicine.

Peter Kaiser: I’m Peter Kaiser. I’m a retinal specialist at the Cole Eye Institute, do a lot with reading centers for clinical trials.

Adnan Butt: So Adnan Butt, biotech analyst at RBC with the Wall Street Perspective.

Mark Humayun: Great. So thank you. For those of you – most of you know them. They’re very seasoned and experienced folks in retina, so it’s good to have you here. The first question we have for you is therapeutic promise of targets and the proposed mechanisms of actions just presented. Did you hear anything that really excited you? I’m sure you’ve heard of these things before. And can you comment on other promising approaches in terms of combination therapy to the current existing anti-VEGFs? So Pravin, I know you’ve been dabbling in this area for some time, so why don’t you comment a little bit about what mechanisms in addition to anti-VEGFs do you think are important?

PD: Well, Mark, I think this is a really exciting time for us.

EC: Pull it a little closer, Pravin.

PD: Yeah. I think it’s a really exciting time for us. I think we’re in a position from a clinical point of view, as well as from an industry point of view where I think we realize that as good as VEGF a suppression is, I think that we’ve reached a ceiling effect of that, and we’re starving for combination therapies. So I think this is a really good time. I’m excited about all the approaches. I think the strategy is kind of interesting, though, which is that there are several great companies going after a particular mechanism of action, you know, Ang and Tie are the ones that come to mind, which validates that target because there are great companies going after that. There are other companies like Ophthea that are going after validated targets, and they’re the only ones that are there such as pan-VEGF inhibition. And there’s advantage to both. I think if you’re the first one there in a mechanism of action that’s unique but validated, I think you would potentially gain a dominant position. So I think that’s an exciting thing on both ends, on both strategies. And finally, there’s a delivery system that is also different. I think I’m excited about the Aerpio delivery system and the fact that there’s a whole new door that opens up with the potential treatment for diabetic retinopathy. And it would seem to me that that delivery is ideal for diabetic retinopathy being sub-q and also going to both eyes. So I think we’re really in for really exciting times, but we’re starved for combination treatment, and I think we’re on the precipice of changing our treatment paradigm.

MB: I have three or four thoughts. I think you can sort of group the improvements into getting compounds that have slightly greater intrinsic efficacy or getting better delivery, or obviously you can obviously you can look at new targets. And we heard today about Tie-2 and a number of other things. But I would say that there’s a lot out there that we could do that we probably aren’t doing as effectively or efficiently as we could be. And one of it is opposed to combination therapy, I think multimodality therapy is either a subset or a superset of combination therapy. For instance, that might be laser plus drugs; it might be surgery ; might be intravitreal injections plus gene therapy. So I think mixing and matching those things works well. Another thing is looking at naturally occurring agents that are more non-specific. I mean steroids are a great example. We product steroids, and yet we still rely upon formulating steroids, and they have a great effect. There’s another class of compounds called crystallines, alpha and beta crystallines. We produce those. They’re close to a super family called heat shock proteins, some of which are chaperones, some of which are solubility factors, some of which are neuroprotective factors. And if you think about it, it’s really interesting when we use low dose laser, whether it’s micro-pulsing or whether it’s very low fluence laser. It looks as though that effect, which may be better than we had ever hoped for, is actually modulated by the expression of HSP70 and other proteins. And some of these are involved in really complex things like folding, tertiary folding of proteins. So that’s another area. And then the third thing I’ll mention is, you know, when you think about it, it takes 2 to tango. When you have essentially a disease that’s mediated by a cytokine or an aberrant protein, it takes a receptor to modulate that effect. These are not toxins. They’re cell modulators. And almost 100% of the time up to this point, at least, we’ve focused on mopping up the agonist or the ligand. And whether it’s a monoclonal antibody or – which is not naturally occurring – or whether it’s something along the lines of a trap, which is the fusion protein with portions of the receptor, nobody’s really looking at binding the receptor. And if you bind the receptor, you can actually inhibit in a non-stoichiometric way – in other words, you can only bind as much agonist as you have binding protein. But if you block a receptor, you can produce an infinite amount of agonist or ligand, and nothing much happens. Now, that’s probably overly simplistic because you have to be sure that, on the one hand, you’re binding the receptor and not turning it on. But there are ways to do that using a structurally modified protein. So I’d say those are the areas that I think are really promising, and that I think have been under-employed, under-investigated up to this point in time.

EC: Don

DD: So I thought the presentations were incredibly interesting. I think the Tie-2 Ang story is really the one that to me strikes me as the closest to making sense, having data, having mechanism that really looks like there’s a drug in there somewhere. I’m very interested in, but a little bit confused by the integrin approach because it seems to do a lot of great things, but it does two wildly different things. I mean it changes your hair color and takes your shoes off, so to speak. And it’s –

EC: What’s wrong with that?

DD: – how those will play in the diseases that we’re treating, some diabetics, we’ve learned that inducing a PVD is a great thing. Years ago, people tried to do it with ultrasound and other things to try to prevent progression of the disease, as you guys well know. But I’m not sure if I right away want a mechanical effect with a biologic effect. But they’ve treated zillions, over a thousand injections. So I’m excited about that. And finally, the approach that I really still think we have short changed steroids. We’ve got a couple of generic and now approved triamcinolone preparations. We have fluocinolone implants and dex implants. And sort of that’s it. But yet I think all the panelists would say if we could recall and bottle in a safe way that wow effect that we got in some of these diseases with the initial intravitreal triamcinolone work, we would really want that in our armamentarium. And yet it’s too much like a baseball bat. So I would hope that we could continue our work of understanding precisely how do steroids produce that effect, and not just giving a class of compound that we found available on our shelves.

MH: Peter, go ahead.

PK: So we’ve been concentrating a lot on combining and making all these diseases one disease that we can treat the same way. But if you really think about it, macular degeneration, diabetic macular edema, even RVO, they’re very different diseases. It just happens to be that anti-VEGF works in all of them because it’s a leakage problem. But if you look at macular degeneration, we‘ve reached a plateau in vision. We’re not really causing the CNV to disappear. We’re decreasing leakage, and that’s where all these other pathways come in. If you look at the angiogenic cascade, every single thing we talked about today is within that angiogenic cascade and which combination is going to work remains to be seen. But what works in AMD may not be the same thing that works in DME. So for instance, a Tie-2 pathway, very important in diabetes, probably a little less important in AMD. But that doesn’t mean the drug shouldn’t go forward. I think in the future we’ll be more talking about these are the DME treatments; these are the AMD treatments.

MB: Right. I think that brings us to the next slide, I believe, about the different mechanisms, if we have that in order. Emmett, think you’re –

EC: I have the controller, sorry. So just sort of to generalize that away from the specific targets that we’ve talked about and more toward what these signaling pathways are doing – are they neuroprotecting, are they causing fibrosis, permeability, angiogenesis? Adnan, maybe I’ll switch back to you and to step away from the assets we heard about. When you think biologically, Adnan, what do you as the analyst, as the scientific mind behind all this, which of the sort of targets and pathways are most attractive to you in AMD versus DME?

AB: So Emmett, if I may, I’ll take that a level higher. So the question I’ll answer is what kind of companies can we fund on the street, what kind of companies are of interest to Wall Street? At this time, both companies that improve vision, improve a clinical benefit, as well as those that improve duration, they’re both fundable companies on Wall Street. So we haven’t really changed the story that much from 2, 3 years ago. Yes, new mechanisms are fantastic, and they’re fantastic because all of us on the sales side, and even on the buy side, are looking to run from Avastin as being the reference drug. So whatever you can give to us and explain to us that will no longer put Avastin as the competitor, that is the fundable asset.

EC: OK. Anybody want to speak to the target science and – let’s stick with AMD first. Is AMD more about dying cells, fibrosis, or persistent fluids, less vessels?

PD: Well, I think we know how patients lose vision. I think we know that over 50% of patients end up having scarring if you follow them long enough. We think we know that in all cases of anti-VEGF-A monotherapy, there’s a period of time after which patients start losing vision gradually. And I hear what you’re saying in terms of the importance of vision, and I truly get that. But the drugs that we have right now are really good drugs. And at the end of the day I think when we develop drugs, I think right now we have a biomarker crisis. Because I think history will show that it’s an absolute coincidence that we’ve developed an anti-permeability drug or drugs, VEGF-A suppression at the same time that OCT developed. At this point, how do we show other than vision? And the vision gain, the margin is really small, in terms of what biomarker do we use for fibrosis? What biomarker do we use for vascular stabilization? And I think that’s the biggest challenge that we have in drug development is that we simply don’t have a good biomarker for these things as we did. And we’ve been spoiled by the biomarker that we have for permeability.

MH: Don.

DD: I’d say that from this list, anti-fibrosis, I think, is my favorite. The neuroprotection one has been incredibly hard to pin down, even in something as simple, for example, as a rhegmatogenous retinal detachment. Still, for all the years that’s 100 years now it’s been successfully treated, there’s still nothing that we do other than put the retina back into its native position that helps the retinal function. We haven’t figured out should we inject glucose in the vitreous or should we hyper-oxygenate people. I mean Dave Zacks and others are looking at adjuvant compounds. So the neuroprotection one is going to be very, very hard to prove. I think anti-permeability and anti-angiogenesis are important, but anti-fibrosis, if you could preserve a retina in a reversible state and not have it become a scar, I think you’d live to fight another day. So I would put that at the top of the list.

EC: I want to ask Adnan, and then I’ll let you comment, Peter, just to follow up. We didn’t list here durability, enhanced durability. So from the Street’s perspective for AMD and DME, what is the OK I’m interested? Is it 3 months, every 3 months? Is it 6 months? What do these companies need to get to as far as dose interval for the Street to say, now you have me?

AB: So six months would be fantastic, but 3 or 4 months is acceptable. And as you know there are phase 3 studies ongoing with that target in mind.

EC: Peter, I’m sorry, I didn’t mean to interrupt you.

PK: That’s all right. I mean I think if you look at macular degeneration in particular, none of our drugs are anti-angiogenic. And that’s the big problem. Maybe that’s why they become fibrotic is because we’re really not reversing CNV. If you look at any of the papers, Marina, Anchor, View, even some of the later Ophthotech, the CNV is not disappearing, and that’s the issue. We need to really cause a CNV to disappear. We don’t have anything that does that yet.

DD: I have a question for Adnan.

EC: I think we’ll go to the next question.

MH: I was just going to comment on that. Peter, I think that’s the holy grail. It’s very difficult to have those blood vessels disappear and yet not cause fibrosis. So how do you do that? I mean how do you get these blood vessels to disappear and not cause fibrosis is really the interesting thing. I think the other that was perhaps left off is inflammation. We didn’t really talk about that. And I think if you blunt the inflammation during this process, you get a little bit potentially less fibrosis as you’re taking away the blood vessel. But Don, you had something?

DD: I had a question for Adnan. So Adnan, if there was a therapy that came out that was not quite as good as first line therapies for AMD, but was given at every 4 months, but not quite as good, how close would it have to be? For example, in the anti-fungal world, when they invented fluconazole compared to amphotericin, I think the figure in the New England Journal trial was it had to be like 75% as effective because it was so simple to give orally. So is there an analytical point at which a sort of a long lasting would be marketable even if it wasn’t quite as good as monthly injection?

AB: So again, I’d go back to 2 points. First, it’s the Avastin being the reference drug as the bugaboo. So maybe you get it used, but it doesn’t quite get used front line. And all of us on the Street, because there is combination data coming by the end of the year, both for Fovista and the Eylea plus PDGFR drug, we’re trying to determine what is that meaningful letter difference. And frankly, probably we’re sort of on the same page as the ASRS survey shows, that you know, 4 or 5 letter, one-line improvement is something meaningful. But where the buy side and sell side are, I think, focusing is something Dr. Boyer pointed to in his presentation, that 3 line gainers, 4 line gainers, maybe there are subsets of data that you can look at which become pain meaningful from an efficacy standpoint. So when you doctors come back to us and tell us something is meaningful to you, it becomes meaningful to us as well.

MH: Go ahead, Mark.

MB: I think I have a slightly different view than the others have expressed. I think the difference between one month and two months and three months and six months are primarily tactical advantages that you have over the short term because there’s always the leapfrog possible. I think what people really are looking for is to cure these diseases. And probably the best way to cure disease is to prevent it, and the second best way is to have an effective treatment. So and there are ways to cure these diseases. Right now we have the technology to do this, whether it’s protein supplementation using things like gene therapy, or whether it’s gene editing. Now we don’t have the luxury of gene editing some of these diseases because they’re polygenic, and it’s just to complicated, and there’s too much work to be done. But we do have within a reasonable sightline the ability to supply proteins at an early stage in these diseases. We know from experimental models – a number of us have been in the lab for years – it’s much harder to cause regression of an established disease such as myocardial contractility or stroke than it is to prevent it. The catch, the bogey is that you have to know who’s going to get it, and then you have to have a therapy that’s safe and effective and relatively noninvasive to do that. But if you could do that, we’d all have that. I’d have that treatment. So I think we have to look a little harder into not treating these disease, but preventing these diseases. And I think it’s within our reach to do that right now.

MH: OK, was it a burning comment quickly or do we move on.

Go ahead Pravin.

PD: A quick comment. It’s not clear to me as we understand the physiology of neovascularization that getting rid of the blood vessel is actually the thing we want to achieve. It’s the wound healing response. And I look at it more as modulating the blood vessel. I think there’s a reason the blood vessel is there in a wound healing response, but we’re just hearing about it. And finally, the other comment that I had is when people ask me all the time, What’s a significant improvement? Four, 5, 2, 3 letters? You know, that’s the median letter gain that you get. It’s a very, very widely distributed disease. The next question ought to be what’s the distribution. If the median is 4, there’s a subgroup of patients that are doing really well, if you can identify that subgroup. At the end of the day the question really should be do you think this study will hit its primary endpoint in the registration. And if it will, it absolutely will be used because our threshold is so low.

EC: So we’ve touched on this question but I’ll let you take the next one if you want. And since we’ve touched on this one. I’m not sure if we want to hit on any other points. This was intended to make the point that we’re going to be seeing this data come out, 4, 5 line mean gain or change, injection burden, we’ve talked about how much is really desired. Anybody want to make one or more comments about how to interpret these data as they come across our desks, whether it’s 4 and a half letter mean gain or –

PK: Well, I think Pravin hit it on the head. Four to 5 letters, it’s a mean. What matters is are you like Alex Rodriguez? So you either hit a home run or you strike out or you or are you more like Ichiro and you get on base all the time. And which one of those 2 you are will matter. So personally, I’d rather see a drug that has a mean where you have a whole bunch of super winners and very quickly can see if you’re not a winner and switch them to something else versus just a minor improvement in vision.

MH: : OK, so let’s go onto the next question, and perhaps maybe the last question is assuming approval of one or more of these discussed drugs as an add on, an adjuvant therapy, how will this impact your patient flow practice? What do you think? How will this play out? Will you be doing two injections on the same say? Would you inject and then perhaps treat with the other drug periodically? How do you see it? Don?

DD: : So I have some views of this. I’m not a fan initially of immediately going to combination drugs. Because I think we’ve only had sort of one class of agents, maybe two if you count steroids. And now we’re going to combine them. We’ll get into the same problems we got into with all the combined glaucoma drugs. I would like to be able to use them individually as we explore. I also think the injections are not going to be a big deal. It often takes me less than 3 or 4 minutes between the time I tell a patient that they’re going to get an injection and then they’re out the door, and I think we’re going to go bing, bing, and I’m not worried about the volume. I think that as the volume increases as it certainly must, we will have ways to deal with it, whether we – and I don’t think it’s going to be multiple visits. I think you’re going to give and then very quickly give a second injection. The 30 minutes is going to be reduced quickly in clinical practice to just a few minutes. That’s my – I think it’s going to be a non-problem I guess is what I’m saying.

MH: Peter?

PK: Yeah, I would agree. I think an injection or two injections, it’s not going to matter. Patients will do it if it improves their vision. That’s the bottom line. Vision trumps all. So if it’s injections, no matter what, we’ll do it on the same day, we’re not going to change how we do it. The thing that’s going to be interesting is if, for instance, the sub-q injection of Aerpio works. That’s something I’ll have to worry about in my clinic, which is great. If the Ora eye drop work and I can give them a prescription and go, Just take this eye drop, that would be great. So there is going to be some differences in the future. But again, what we’ll use won’t matter except for the vision. So whatever is the best vision, that’s the drug we’re going to use.

MH: Pravin?

PD: I think at the end of the day, efficacy is king. Efficacy always wins. And I’m old enough to remember when we sat there and said, Who would ever take a needle in their eye once every 4 or 5 months, and look at what we’re doing now. That’s number one. Number 2 is I really look forward to having a portfolio of agents, and the beauty of this is that none of these mechanisms of action are exclusive to another. I hope they all work, and I hope we’ll use them all, and I look forward to having a portfolio where we’ll be able to truly individualize and customize care. And I think that’s what it will come down to: the practice in 2020 where we’ll see patients and we’ll be able to individualize treatment and customize their selection of combination therapy. I think that’s the ultimate goal.

EC: Thank you all. We’ve run a bit late, and I have to thank my panelists. Thank you for initiating this long and complicated discussion.


Get the Latest News, Podcasts and Videos.

Scroll To Top