His Yates Lab at UVA is a leader in retina research.

Paul A. Yates, MD, PhD, associate professor of ophthalmology at the University of Virginia (UVA), first came to our attention as the initiator of a large-scale, current clinical trial to employ the FDA-approved oral rosacea drug Oracea (doxycycline; Galderma) as a potential therapy for geographic atrophy. What struck us most was that Dr. Yates believes the study can still be a success even if it does not produce a commercial product. In other words, his primary goal is to advance the knowledge of this disease so that other researchers can use his work as a stepping stone to effective treatments. Until now, Dr. Yates has maintained a low profile in the retina community, but his compelling insights merit recognition.

Q. When did you start your lab at UVA and what were your primary interests and goals in doing so?

A. My lab was established in 2008, shortly after I arrived in Charlottesville, Virginia. The goal of the lab, like that of most clinician-scientists, is to develop new insights into retinal disease and hopefully translate that knowledge into improved therapies for my patients. While anti-VEGF injections have demonstrated the enormous impact that eye research can have on patients’ lives, unfortunately many patients arrive too late in their disease process or remain unresponsive to these treatments to ultimately preserve vision.

With that in mind, my lab has tried to focus at the polar ends of retinal disease by developing new approaches for imaging the retina to identify patients earlier in their disease process when lifestyle changes can still impact final vision outcomes, as well as more bench-level research on mechanisms of end-stage AMD and diabetic retinopathy to identify novel drug targets for patients with advanced disease.

Q. In what areas of your research do you believe you are making the most progress toward your goals?

A. The dream of any clinician-scientist is to see their research efforts eventually help patients, and I will always measure my personal success by the number of people on whom we can have an impact. Thus, I am most excited to be part of the TOGA (Treatment with Oracea for Geographic Atrophy) study, because it is a chance to finally determine if this therapeutic approach can slow the progression of AMD. The study was enabled by more than 40 like-minded research sites across the country that generously donated their time, a gift from the Manning Family Foundation, and study drug donation from Galderma. I am enormously grateful for this opportunity.

We have also found success with the imaging technology developed in my lab, having founded a start-up to commercialize affordable widefield retinal imagers for screening of premature infants for ROP and adults for diabetic retinopathy. These cameras, in visualizing the entire retina, should enable improved detection of retinal disease to prevent vision loss. A key relationship with Olympus, as well as other ophthalmic industry partners, has provided us access to state-of-the-art optics and imaging sensors that make our devices possible. While we often ascribe accomplishments to individuals, in fact the coordinated effort of many is needed to fully realize new therapeutics and technologies.

Q.  You are currently conducting a large-scale clinical study of more than 200 patients, using the oral drug Oracea, which is approved for the treatment of rosacea, as a potential therapy for GA. Why did you choose Oracea for this study and what do you hope to accomplish?

A. Tetracycline derivatives, such as Oracea, have long been used for treatment of ocular rosacea and blepharitis, with efficacy related to their anti-inflammatory, rather than antimicrobial, properties. Mechanistic studies have demonstrated that the anti-inflammatory properties result from their inhibition of matrix metalloproteinases and caspases, generation of reactive oxygen species, and complement activation. These effects in the eye and brain appear to occur in part because they inhibit microglial activation to a proinflammatory M1 state.

We know that both microglial activation and many of these proinflammatory pathways contribute to the progression of AMD, suggesting that doxycycline may be able to impact this disease. Doxycycline was chosen specifically because, unlike other tetracycline derivatives, it has a well-known dosing split between its anti-inflammatory (low-dose 20 mg twice a day) and antimicrobial (high-dose 100 mg twice a day) activities. Thus, we could avoid creating potential antibiotic resistance in our elderly study patients, which is important, given the length of time they would be on this treatment.

Oracea is an extended-release version of low-dose doxycycline, which Galderma generously agreed to donate to our study. As with most studies of dry macular degeneration, we are similarly looking for a potential impact on the rate of growth for end-stage GA lesions. While there is some debate within the field about whether this is the most appropriate metric, or stage of disease at which to intervene, it is an agreed-upon endpoint for demonstrating treatment efficacy.

Q. You have said that a study such as the one you are conducting does not have to result in a commercial product to be considered a success. So would it be enough just to move the needle on what we know about GA and what types of therapies may be able to combat it?

A. As with most chronic diseases, AMD is very likely multifactorial, and as such may not be as amenable to a single-target, anti-VEGF-like approach. Animal models for AMD and genome-wide association studies have helped identify many potential avenues for therapy. The problem is that it is difficult to know a priori the relative contributions of each pathway, and as suggested by the MAHALO study, therapy may be patient specific.

It is informative to probe several disparate treatment approaches to better understand the impact on disease in patients. Acucela unfortunately did not demonstrate efficacy for emixustat, but it nevertheless gave important insight into visual-cycle modulation effects on AMD. Similarly, GlaxoSmithKline did not show efficacy for GSK933776, but its approach to inhibit beta amyloid was sound, based on animal models.

To paraphrase Thomas Edison, we shouldn’t view these trials as failures, but rather as yet another method that doesn’t work, which will provide more information for researchers who follow. Although I would love for our trial to be a success, and it will not feel great if it too fails to meet its endpoints, we will likely learn something about microglial inhibition and its role in AMD in the process. As I tell my AMD patients, I don’t really care who develops the first successful therapy for dry AMD, I just care that we eventually have one so that I have something to offer my patients.

Q. You follow just about every credible research initiative currently going on in GA. How would you assess the progress being made, if any, and what approaches appear to have the best chance of success?

A. Progress has been phenomenal, and the drug pipeline is so full right now that you have to believe that one of the candidate drugs will demonstrate efficacy. Clearly the phase 2 evidence would suggest that either lampalizumab (Genentech) or brimonidine (Allergan) have the best chance of success among drug candidates for slowing progression of GA. For AMD there were disappointing results recently in phase 2 of the CAPELLA trial, and there were encouraging results for the phase 2 Fovista study, but it did not ultimately pan out in phase 3. No single drug, target, or approach (ie, MC-1101, fluocinolone, complement inhibitors, CD59, metformin, micropulse laser, light mask therapy) can be ruled out.

It’s inspiring that so many people are investing so much effort into going after this problem. Both my hope and my worry is that either lampalizumab or brimondine prove efficacious. It’s my hope because we need a therapy for our dry AMD patients. It’s my worry because if these therapies become standard of care, the cost of further drug trials starts to become prohibitively expensive, and that may collapse the pipeline pretty quickly.

Part of the reason we initiated our trial when we did was that the financial bar for entry was much lower. Our trial with almost 300 patients will cost less than $500,000, and that opportunity will be forever lost when any new therapeutic has to be directly compared to a $2,000 biologic injected monthly.

Q. For-profit ophthalmic companies probably wouldn’t even begin a clinical trial if they didn’t see the possibility of a commercial product eventually coming out of it. Do you think some trials (especially large studies based on genetic factors) should be underwritten by the NIH, universities, or private foundations because they won’t be done by for-profit companies?

A. Despite their size, most large pharmaceutical companies remain relatively risk averse. With cost to market now ranging between $1 billion and $2 billion for a new drug, one can begin to understand the reasoning, because even for a major pharmaceutical company, that is a sizeable bet to place. Both public and private foundation funding will continue to be an important part of our drug-development pipeline, because the tolerance for risk is higher, shorter-term financial goals do not have to be met, and there remains incentive for smaller companies and individuals to lay much of the groundwork for new innovations with reasonable rewards for doing so.

Q. What are the challenges with securing capital?

A. The reality is that the costs of performing a major pharmaceutical trial (our study excepted), especially for injectable biologics, makes it cost prohibitive for any small pharmaceutical company to go it alone. There has been a trend for some of these smaller ophthalmic start-ups to offer an IPO to fund their clinical development, but as we have seen recently, a single failure or even an apparent setback can be devastating. The more we assume that for-profit companies will pick up the slack in the growing public funding gap, the more we put additional therapeutic advances at risk.

For-profit companies have an important function in taking the best academic ideas that have been partially de-risked by public funding and seeing if they have merit. Our current limits on growth of public funding are likely to have significant consequences downstream as more researchers, like myself, leave to find more fertile ground.

Q. What have the clinical trial failures taught us? Why do we see such disparities between phase 2 and phase 3 data?

A. The recent failure of Ophthotech and Regeneron trials have no big-picture implications to me beyond the fact that drug development is hard and not for the faint of heart. Half of all phase 3 trials fail despite promising phase 2 data.

The failures are very likely in clinical trial design, specifically endpoints and patient selection, as well as a failure to correctly determine the optimal dose. But to call them failures really paints them in an unfair negative light. It is unlikely that perfect trial design would result in 100% success moving from phase 2 to phase 3.

We make many, perhaps overly optimistic, assumptions about equivalence of treated and placebo patient populations with phase 2 trials. There may necessarily be undetectable disparities between control and treatment populations that if we could appropriately model in an ANOVA would completely explain the observed results. It’s also possible that attrition and rater bias or even poor statistical methods may creep into study design and data analysis. But, perfection in small study designs is impossible, and a statistical test is just that — a probability.

Given we accept P<.05 as our threshold for efficacy (and some trials go into phase 3 even without significance in phase 2) that implies 1 out of 20 positive results are due to chance alone. To me, the only shocking thing is that these trial outcomes are called shocking, when in actual fact they are more the norm.

Q. Do the 2 companies that already have FDA-approved retinal therapies (Genentech and Regeneron) have advantages in retaining their dominance into the next generation of retinal drugs?

A. Regeneron and Genentech have the resources, pipeline, and development processes to continue to rule the injectable ophthalmic biologics space for at least the next decade. This is well reflected in Genentech’s CrossMab technology, currently in trial to provide dual inhibition of VEGF and Ang2. Similarly, Regeneron has its TRAP technology and VelocImmune mouse, which can significantly speed development and evaluation of potential therapeutics. Ultimately, trial outcomes, and demonstrated efficacy, are what will drive market share.

Q. Now that we have had successful drugs for retinal diseases for a decade, are new trends emerging as to targets for drug development?

A. Biologics will continue to rule the retina space but the opportunity exists for increasing our efficiency in and patient acceptance of intravitreal injections. Use of long-term release depots or adenoviral delivery methods holds promise to reduce overall treatment burden, and I see that approach only growing over the coming decade. The antiangiogenic pipelines will likely continue, though with the recent failure of PDGF, a subsequent failure with Ang2/Tie2 may significantly temper development effort. I remain optimistic about Ang2 as a target, but I’ll admit that is perhaps just a personal bias.

There is also potential for the other major player in the vascular unit, the pericyte. The trophic machinery of perivascular cells for support of a healthy vasculature has not been well targeted thus far, though in part this is due to a lack of full understanding of these cells and a noninvasive means to assess function. Our group has recently shown through lineage markers that vascular smooth muscle cells throughout the retinal microvasculature have mesenchymal stem cell properties. Manipulation of these properties carries the promise of supportive, regenerative, and even potentially deleterious effects on retinal health. The promise of stem cell therapeutics for the eye remains, but there are many delivery, dosing, and longevity issues that will take a while to optimize before this therapy becomes routine.

Q. If we could look ahead 5 to 10 years, how will retina specialists be treating retinal diseases? Will it be anti-VEGF monotherapy, combinations, sustained-release platforms, or even gene or stem cell therapy?

A. I have no doubt that anti-VEGF in some form will continue to play a large part in our therapeutic approach to most retinal disease. There will continue to be combinatorial approaches to improving treatment efficacy, though here the bar to cross will be much higher than the pre-anti-VEGF time period.

The question with the recent platelet-derived growth factor trial failures is whether it was a failure of the drug to provide additional efficacy or of the investigators to select the appropriate subgroup of patients who would be most responsive. I foresee a time when we can profile the proangiogenic, proinflammatory mix for individual patients to achieve the best outcomes. Essentially we may well turn retinal disease into a version of the glaucoma market. I expect retinal specialists’ armamentarium to continue to grow and retina will continue to be an exciting field. RP