Michael Trese, MD: Vitreomacular adhesion (VMA) is a condition when the vitreous gel adheres in an abnormally strong manner to the retina. VMA can lead to vitreomacular traction (VMT) and subsequent loss or distortion (metamorphopsia) of visual acuity, a condition known as symptomatic VMA. VMA occurs in the context of an incomplete or anomalous posterior vitreous detachment (PVD) and is linked to several retinal disorders including macular hole, epiretinal membrane (ERM), neovascular age-related macular degeneration (AMD), diabetic macular edema (DME), retina vein occlusion (RVO), and retinal tears and detachment.1

SYMPTOMATIC VMA AND ANOMALOUS PVD

Dr. Trese: Dr. Brown, can you explain the process by which VMA and anomalous posterior vitreous detachment (PVD) occurs?

David M. Brown, MD: The simplest way to explain the process of VMA and anomalous PVD is how I explain it to my patients. I tell them that there is a gel inside the eye and that as a person gets older, the gel degenerates and clumps up creating floaters. Eventually the gel structure cannot keep its shape as a Jello ball, and it collapses in on itself. The vitreous gel is connected to the retina similar to Velcro and the adhesion is usually strongest at the optic nerve and at the fovea. When it pulls on the optic disc, it does not create much of a problem, but when it pulls on the fovea, the most delicate part of the photoreceptors and the thinnest part of the retina, it can create a macular hole, which requires surgical intervention.

Peter K. Kaiser, MD: Most of us will eventually develop a posterior vitreous separation as a result of age, which involves both liquefaction of the vitreous as well as release of the vitreous from the retina. If the vitreous liquifies and releases uniformly and in a synchronized manner, the result is a successful PVD; however, if VMA develops in the foveal region, the result can include macular hole, macular traction, vitreomacular traction syndrome (VMT), and in patients with diabetes, posterior hyaloid traction.

Pravin U. Dugel, MD: We are beginning to understand that a PVD is not as simple as we once thought it was. It is important to note, as Dr. Kaiser stated, that there are two distinct steps that occur: liquefaction, or synesis, and separation of the interface, or syneresis, and if that two-step process does not occur in proper synchrony and completion, anomalous PVD can result, leading to VMA.

Years ago, Jerry Sebag, MD, talked about a unifying concept in anomalous PVD.1 A VMA resulting from anomalous PVD can go one of two ways. The vitreous can actually split to form a schisis and a partial thickness PVD, resulting in traction. If the traction goes outward, a macular hole may result. If the traction goes inward, an epiretinal membrane (ERM) or macular pucker may form. A full thickness separation of the vitreous may occur, but it may be caught up somewhere. If it is caught in the posterior pole, and posterior traction occurs with peripheral separation, macular traction may result. Macular traction can cause vitreomacular traction syndrome, but it is also linked to many other conditions such as neovascular AMD, diabetic retinopathy, DME, and RVO. If traction occurs at the optic disc, vitreopapillary syndrome, and ensuing disc edema, can occur, and if there is posterior separation but peripheral traction, retinal tears and retinal detachment can occur (Figure 1).

It is interesting to see how VMA resulting from anomalous PVD is involved in so many diseases—I think that we are gaining more understanding because our imaging techniques are now married with what we know about the pathophysiology of this process.

Mark Humayun, MD: In the era of optical coherence tomography (OCT), which allows for easier detection of macular adhesions, what is the definition of symptomatic VMA? Perhaps we can see distortion of the retina on our OCT scans, but the patient may or may not attribute their symptoms to this condition. It is interesting, particularly in an age where we are able to see these adhesion areas much better.

Dr. Trese: The diagnosis of PVD, which we all have on our diagnostic sheets, is really an exclusion of a retinal tear. How many times have each of us seen a patient who presents with flashes and floaters and diagnosed PVD, only to have the patient come back 2 years later with flashes and floaters, leading to the realization that the rest their vitreous has probably separated? As previously stated, there are two separate concepts that are involved in the change in the vitreous—liquefaction and separation. Liquefaction is a core change and separation is a vitreoretinal juncture change. If the core change occurs without the juncture change, or the two do not take place relatively close in time to one another, anomalous PVD and subsequent VMA leading to traction and associated symptoms can occur.

Dr Kaiser: Anomalous PVD is really any time liquefaction and separation do not occur at the same time. If the liquefaction develops without the release, symptomatic VMA or a tear in the periphery can result.

Dr. Trese: If you look at the dispase study, which evaluated this agent that liquifies the vitreous without causing vitreous-retina interface separation, the retinal detachment rate was approximately 5%, which highlights the danger of inducing liquefaction without manipulation of the vitreoretinal juncture.2

Dr. Brown: Both liquefaction and syneresis occur naturally as the eye ages. Disease processes, however, such as diabetes and AMD, cause exudation and make the retinal interfaces thicker and harder to break. Thus, there may be a higher incidence of anomalous PVDs as the gel needs to pull away in these eyes, resulting in more pathology.

Dr. Trese: In the case of diabetes, hyperglycemia alone increases laminin and fibronectin, two of the three components in the vitreoretinal juncture as well as the glomerular extracellular matrix, tightening the vitreoretinal juncture and reducing the filtration rate from the kidneys.

Dr. Humayun: There are many different ways to think about VMA. Most of us know that the macula is the area we are trying to protect from developing neovascularization, hemorrhage, or macular edema. We also know that vitreous is adherently attached to the optic nerve and along the vessels and can create VMT. This interface is complex—it can have both fibrous and cellular components. As we better understand the retinal interface, it is becoming clear that it plays a role in many retina diseases, particularly when there is traction on the retina.

It would be advantageous to use the tools we currently have to perform an optical-imaging biopsy with OCT under dynamic conditions to better understand the mechanical functions.

PHARMACOLOGIC VITREOLYSIS

Dr. Trese: How would you define pharmacologic vitreolysis?

Dr. Brown: Pharmacologic vitreolysis or pharmacologic vitreodynamics is the process by which an enzyme is used to mechanically change the vitreous. This involves two distinct mechanisms. One mechanism is liquefaction of the core vitreous and the other is degradation of the vitreoretinal interface. By doing this we change not only the physical traction where the “Velcro” is trying to separate from the retina, but also changes in the milieu with the oxygenation and chemical balance that go along with this separation of the vitreous from the retina.

Dr. Trese: What is the role of the vitreous in disease states such as AMD, DME, and RVO?

Dr. Dugel: It may be too early to tell, but there are some articles that are coming out that are compelling. A study from Susan Binder, MD's, group showed that PVD may be protective against AMD, whereas VMA may promote exudative AMD.3 Another earlier paper from Dr. Binder's group demonstrated that persistent PVD may be a risk factor for exudative AMD from inflammationinduced vitreoretinal traction.4

The role of the vitreous in disease states like AMD is interesting. Obviously, we have an unsustainable model in our current monotherapy treatment of AMD with injections of anti-VEGF, and I think that we can all agree that combination therapy is an improvement on monotherapy. However, we have just recently started talking about manipulation of the vitreous and how this may play an important role in disease management.

Your group published an animal study regarding the effect of bevacizumab on vitreous detachment.5 A study such as this, which evaluates the effects of pharmacologic agents on vitreous manipulation, may have important implications.

Dr. Kaiser: It is difficult to grasp the extent of the vitreous' role once a patient has developed choroidal neovascularization (CNV) and AMD. There is clearly a strong association, as evidenced in the literature.3,4,6,7 The question that remains, however, is if CNV is already present, would inducing vitreous separation alleviate symptoms or enhance the effect of an anti-VEGF agent to decrease the number of injections needed?

Dr. Trese: Dr. Dugel, can you describe the enzymatic agent, ocriplasmin (ThromboGenics), formerly known as microplasmin that has been under study for pharmacologic vitreolysis for symptomatic VMA?

Dr. Dugel: Ocriplasmin is a truncated form of human plasmin. Ocriplasmin is not a new drug—it has been in development for many years and there are excellent phase 1 and phase 2 studies that were previously encouraging. The phase 3 studies, which were two large multicenter, prospective, randomized trials, were remarkably consistent in regard to the safety and efficacy outcomes.8,9 As far as clinical use is concerned, this drug was tested in a diverse population of patients with diseases such as VMT, macular hole, and ERM (See “Key Published Literature”). The applicability, however, will extend far beyond these diseases. It will go to other diseases where VMT may be involved such as DME and neovascular AMD. It will also extend to the combination of ocriplasmin with other drugs such as anti-VEGF agents; there is now evidence that this drug may make anti-VEGF treatments more effective when used in combination.5 Finally, I think this drug will also be used as an adjunct to surgery. The potential of injecting this drug the day before surgery and making a very complicated case much easier is intriguing. In general, I think that we have a great deal of evidence regarding this drug but the potential clinical applications of pharmacologic vitreolysis may be enormous.

POTENTIAL APPLICATIONS FOR VITREOLYSIS

Dr. Brown: One of the unmet medical needs in terms of classic VMT is for patients with impending macular hole but relatively good visual acuity, placing the surgeon on the fence as to whether to operate sooner or later. To have an agent for vitreolysis would help a fair number of our patients avoid surgery completely.

Dr. Kaiser: You could take it even one step further. Patients with so-called stage zero macular holes where the fellow eye has a gull-wing shaped vitreous coming down to the fovea but no hole and no true traction may be a higher risk for developing a bilateral hole some time in the future.10 Should we treat these patients with a vitreolysis agent as prophylaxis for a potential future problem?

Dr. Trese: Gull-shaped retinal detachments create a large amount of traction. Lois Smith, MD, PhD, told me about an individual in the lab at Children's Hospital Boston who has demonstrated that isolated traction on Muller cells can generate VEGF. I am not sure what role vitreoretinal adhesion plays in AMD, but I agree that it is most likely early in the process. Once CNV is present, we need to study if cleaving the adhesion will help.

Dr. Dugel: I also see potential for vitreolysis agents as adjuncts to surgery. Some of the hardest cases we see are complicated tractional retinal detachments in diabetics and retinal detachments in pediatric patients. If we can inject a vitreolysis agent a day prior to surgery, it might make the the surgery easier and more successful.

Dr. Brown: There is postulation that schisis pockets can act as a sink for VEGF. We know that after a PVD there is more oxygenation, so a PVD would be ideal for every patient with diabetes. Even when the glycosalated hemoglobin is causing vascular damage, you want more oxygen for the retina and a better escape for any ischemic factors. I do not see any reason not to induce a PVD as long as I have an agent that is effective and safe.

Dr. Kaiser: Assuming we had a drug that is effective and safe, it would be best to induce PVD prior to the development of exudation and protein release that increases adherence, because after these occur, it will become more difficult to release the junction. So I think you can argue that it is reasonable to inject with ocriplasmin when mild changes become apparent to prevent future complications. Dr. Dugel: Because we know a lot more about the role of the vitreous in diabetes than we do in AMD, we are more apt to argue for early injection. The safety profile of ocriplasmin was shown to be excellent in the phase 3 trials.8,9 In fact, there were more cases of retinal tears and detachments in the placebo group than in the drug group, so these were most likely complications of vitrectomy.

Dr. Trese: Let's say that sometime in the future you have a patient with a 9-year history of diabetes who has a couple of dot and blot hemorrhages and no PVD on OCT or ultrasound. What do you do if, after injecting ocriplasmin, a PVD does not result? Peter Stalmans, MD, presented data from the phase 2 ocriplasmin study on a group of patients with diabetes. He injected the enzyme in these patients, many of whom had proliferation and had undergone laser and approximately 11% to 14% cleaved. In my opinion, this is a miracle because there was no case selection.12

If you did not have a PVD in such a patient, would you reinject?

Dr. Dugel: I would reinject. Regarding the data to which you refer from the phase 2 trials, there was one cohort of nine patients for whom reinjection was allowed. Although the group was small, the rate of PVD after reinjection was almost 60%.12 Given the very strong safety profile of ocriplasmin in over 800 eyes, I would not hesitate to inject two or three times.

Dr. Kaiser: The phase 3 study did not look at reinjection, but I think that it is important to consider that maybe the first injection is just not getting the drug to the right location. With a second or third injection, you will certainly increase the likelihood of getting it to the right place and being able to cleave the junction. If and when ocriplasmin is approved by the US Food and Drug Administration, a good investigator- sponsored trial might include an analysis of reinjection data.

Dr. Trese: Another scenario where vitreous manipulation may be justified is in cases of capillary dropout in retinal vein occlusion. Injecting ocriplasmin could help to re-establish circulation in some of these cases.

Dr. Dugel: Our whole way of thinking of neovascularization might change if we could use vitreous manipulation to affect circulation. Currently, in patients with ischemia form any cause, we have no effective treatment. It would be very valuable to have a drug that could increase circulation by chemical vitreous manipulation. This is plausible: there is good evidence, originally from Stefansson, that mechanical vitreous removal increases oxygenation.13

Dr. Trese: Imagine that you are examining a patient who has VMT due to VMA without a membrane, a macular hole with less than a 1,500-μm attachment, and lattice degeneration in the periphery. How do you handle this case?

Dr. Brown: Because the induction of a PVD might very well increase the risk of a peripheral tear in this patient with peripheral lattice, I would probably demarcate the peripheral lattice degeneration lesions with thermal laser at least 3 or 4 days prior to injecting an agent to induce pharmacologic vitreolysis. In the ocriplasmin clinical trials at our site, I did not have any cases of retinal tearing, but we did not have many patients with peripheral pathology.

,strong>Dr. Trese: One of the first patients treated in the clinical trials in Europe did have a retinal tear. The clinician who was treating the patient did not see any sign of lattice degeneration in the periphery. However, after being treated with laser, they did well. Although the tear rate was not high in the clinical trials, it is something that we need to think about.

Dr. Kaiser: Although I do not consider lattice degeneration a contraindication for ocriplasmin injection, it is important to discuss the risks of retinal tears with your patients. It is important to conduct a thorough exam prior to considering this type of injection, and if lattice degeneration is detected, the option of treating with laser prior to injection vs just injection should be discussed. Additionally, patients should be counseled on symptoms of retinal tear, should it occur.

MAXIMIZING THE MECHANICAL AND BIOCHEMICAL ACTIVITY OF OCRIPLASMIN

Dr. Dugel: The size of the VMA is directly related to how well ocriplasmin works. Additionally, the location of the injection is important. If it is injected away from the pathology, it will be hard for the drug to reach the location of the adhesion.

Dr. Trese: If you look at the 250-μm adhesions from the MIVI-TRUST phase 3 clinical trials for ocriplasmin, there were approximately 60% of these adhesions that cleaved.8,9 For those that were 1,500 μm or larger, the numbers that cleaved were small. How do you assess this clinically? First, as Dr. Dugel said, delivery of the enzyme is critical to get the ocriplasmin to where it can work.

We presented a poster at the Association for Research in Vision and Ophthalmology 2011 meeting that shows a technique that attempts to achieve a larger area of drug from the same bolus. We found that when injecting India ink with a 30-gauge quadraport needle, as compared to a standard 30-gauge needle to inject India ink, the ink spread more quickly and to a wider area, whereas the ink accumulated around the tip of the standard 30-gauge needle.11

In another group of animals injected with ocriplasmin and plasmin using a standard 30-gauge needle vs the quadraport 30-gauge needle, scanning electron microscopy at 1 week showed that there was no residual enzyme on the internal limiting membrane with the quadraport needle.11

Dr. Dugel: Achieving needle delivery that will allow for selective infusion of the drug to the site where it is needed is a challenging task. Approximately 14% of patients and 20% of patients without an ERM treated with ocriplasmin in the pivotal phase 3 trials achieved the secondary endpoint of total PVD induction.8,9

Dr. Humayun: Mechanically speaking, the idea of injecting in different places, lifting the attachment, and eventually making the entire sheet come off seems to make sense.

Dr. Brown: So you are saying is that after injecting ocriplasmin in several locations, if you lift up one edge and you move it back and forth, this may be of benefit and a more gentle vitreolysis?

Dr. Dugel: Actually, sequential injections may be easier. We have some data on this from the phase 2 trials regarding these from one cohort in which this was allowed, and it was shown that sequential injections my increase the rate of total PVD without having to move the patient's eye.

Dr. Trese: When discussing how to maximize mechanical and biochemical activity, reliable imaging may enhance the ability to target exactly where you want to inject the enzyme. The dynamic SD-OCT imaging that we have with the Spectralis (Heidelberg Engineering, Carlsbad, CA) offers a better image of the vitreous so that we can better understand its relationship to retinal disease.

We have put together a procedure room to perform the injections of ocriplasmin that allows us to do the injection through an operating microscope. It would be ideal to be able to do these injections in an OCT-guided manner.

Dr. Kaiser: You already have half of the system in place. In the future, microscopes with built-in OCT systems will be available. Currently, the Bioptigen Spectral Domain Ophthalmic Imaging System (SDOIS; Research Triangle Park, NC) has a mount for most microscopes, and Carl Zeiss Meditec (Dublin, CA) is working on a version of microscope with OCT that is built in. The ability to see underneath the vitreous during injection is imminent.

Dr. Brown: What do we know of the diffusion of ocriplasmin?

Dr. Trese: The way that ocriplasmin works is that it takes one bite and then it either binds to substrate or consumes itself. So if you inject it in a large bolus through a standard needle it consumes substrate and then itself. If you use a multiport needle, however, such as the quadraport needle that was used in our study, the chances of getting more substrate are higher.

CHANGING TREATMENT PARADIGMS

Dr. Trese: This has been an interesting year in retina. We learned the results of the Comparison of AMD Treatments Trial, which showed that an inexpensive anti- VEGF agent, bevacizumab (Avastin, Genentech) is noninferior to a very expensive anti-VEGF agent, ranibizumab (Lucentis, Genentech). We also have results from the two large clinical trials showing the safety and efficacy of VEGF trap (Eyelea, Regeneron), although we do not know how this drug will be priced when and if it is approved.

How do you think that physicians feel about something that takes away few of the surgical procedures that they enjoy performing and that may represent another expensive drug to handle in the office?

Dr. Kaiser: As physicians, we all have the same goal: to treat patients in the best and safest way we can. Vitrectomy is certainly an excellent way to fix symptomatic VMAs, but if a safer, quicker way exists to do it, it is better for the patient.

Dr. Dugel: I agree. However, there are many instances where surgery is still indicated. In the clinical trials for ocriplasmin, there was almost a 60% closure rate of macular hole with a single injection if the hole was 250 μm or smaller.9,10 But if in the case of a large macular hole with no VMA, there is no reason to inject ocriplasmin; we still need to perform surgery.

Another consideration is that this drug may actually enhance our surgery. I can picture using this drug for complicated traction detachment or for a young patient who requires a PVD, so I do not think that pharmacologic vitreolysis is going to take away from our surgery. Also, in patients with ERM, ocriplasmin will help to resolve part of the problem; symptoms associated with VMA leading to antereoposterior traction. We will still have to perform surgeries to relieve the tangential forces exerted by an ERM. In fact, I think ocriplasmin will be a safe and effective alternative for patients who currently have no treatment options, patients with VMA and early symptoms, and will be an effective adjuvant for difficult surgical cases, making us better surgeons. I do not see push back of this drug from surgeons. On the contrary, I predict surgeons will see the potential beneficial effect of this drug for our patients in the office and in surgery and will welcome access to it.

Dr. Humayun: In ophthalmology, when a noninvasive option becomes available we typically intervene earlier. Ocriplasmin will allow us to treat patients in whom there clearly is pathology and who are symptomatic but for whom surgery is not indicated because of the poor risk/benefit ratio. I do not think that there is much overlap with surgery, particularly when you are talking about earlier disease.

Dr. Trese: I think that eventually, criteria for early intervention for symptomatic VMA with ocriplasmin will be developed. I imagine that the American Academy of Ophthalmology will produce guidelines based on highresolution OCT imaging to help physicians in using this drug effectively.

SUMMARY

Dr. Kaiser: The ability to treat symptomatic VMA, particularly early macular holes, with pharmacologic vitreolysis offers an efficient and safe way to improve our patients symptoms without requiring surgery or facedown positioning. Moreover, the therapy may help us if surgery is required in the future, as the vitreous is more liquefied and easier to peel.

Dr. Brown: The addition of ocriplasmin will be a welcome addition to the armamentarium of vitreoretinal surgeons. The agent will be first used in patients with impending macular holes and small holes with persistent VMT at the edges. If multiple injections (or novel injection techniques) can result in reliable pharmacologic vitreolysis, I can easily see the agent being used routinely before vitreoretinal surgery and possibly as a prophylactic agent in diabetic patients prior to the development of proliferative disease.

Dr. Humayun: As the field of enzymatic vitreolysis continues to advance and FDA-approved methodologies become available, there is no doubt that such an approach will be used to treat symptomatic VMA.

Dr. Dugel: There are three forces that I believe will come together to make ocriplasmin a very valuable drug for the retina specialist. The first force is the advent of technological advances with OCT that allow us to easily visualize VMA. The second force is that all retina specialists are adept at performing an intravitreal injection. The third force is that we currently employ a watch-and-wait strategy for many of our patients due to the lack of options for a safe, noninvasive procedure. If ocriplasmin becomes available, I believe that retina practices are prime to adopt this agent for pharmacologic vitreolysis.

  1. Sebag J. Anomalous posterior vitreous detachment: a unifying concept in vitreo-retinal disease. Graefes Arch Clin Exp Ophthalmol. 2004;242(8):690-698.
  2. Tezel TH, Del Priore LV, Kaplan HJ. Posterior vitreous detachment with dispase. Retina. 1998;18:7-15.
  3. Robison CD, Krebs I, Binder S, et al. Vitreomacular adhesion in active and end-stage age-related macular degeneration. Am J Ophthalmol. 2009;148(1):79-82.
  5. Krebs I, Brannath W, Glittenberg C, Zeiler F, Sebag J, Binder S. Posterior vitreomacular adhesion: a potential risk factor for exudative age-related macular degeneration? Am J Ophthalmol. 2007;144(5):741-746.
  6. Goldenberg DT, Giblin FJ, Cheng M, et al. Posterior vitreous detachment with microplasmin alters the retinal penetration of intravitreal bevacizumab (Avastin) in rabbit eyes. Retina. 2011;31(2):393-400.
  7. Nomura Y, Ueta T, Iriyama A, et al. Vitreomacular interface in typical exudative agerelated macular degeneration and polypoidal choroidal vasculopathy. Ophthalmology. 2011;118(5):853-859.
  8. Stefánsson E, Geirsdóttir A, Sigurdsson H. Metabolic physiology in age related macular degeneration.Prog Retin Eye Res. 2011;30(1):72-80.
  9. Jumper J, Pakola S. The MIVI-007 trial. Phase 3 evaluation of single intravitreous injection of microplasmin or placebo for treatment of focal vitreomacular adhesion. Paper presented at: the American Society of Retina Specialists; August 31, 2010; Vancouver, BC.
  10. Packo K, Pakola S. The MIVI-006 trial. Phase 3 evaluation of single intravitreous injection of microplasmin or placebo for treatment of focal vitreomacular adhesion. Paper presented at: the American Society of Retina Specialists; August 31, 2010; Vancouver, BC.
  11. Chan A, Duker JS, Schuman JS, Fujimoto JG. Stage 0 macular holes: observations by optical coherence tomography. Ophthalmology. 2004;111(11):2027-2032.
  12. Trese MT, Asami T. A quadraport needle gives better spread of drug during intravitreous injection. Poster presented at the Association for Research in Vision and Ophthalmology annual meeting; May 1-5, 2011; Fort Lauderdale, FL.
  13. Stalmans P, Delaey C, de Smet, M, van Dijkman E, Pakola S. Intravitreal injection of microplasmin for treatment of vitreomacular adhesion. Results of a prospective, randomized, sham-controlled phase II trial (The MIVI-IIT Trial). Retina. 2010.?30(7):1122-1127.
  14. Stefánsson E. Ocular oxygenation and the treatment of diabetic retinopathy. Surv Ophthalmol. 2006;51(4)364-380.