Specialists weigh indications, outcomes of vitreomacular interface disease treatment

Inroads are being made in the nonsurgical treatment of vitreomacular disease. Ocriplasmin is a game-changer in the treatment of vitreomacular adhesion, vitreomacular traction and associated macular holes, according to some retina specialists.

“Ocriplasmin is a new drug that is initiating a new paradigm in treating vitreomacular adhesion. Vitrectomy used to be the only treatment to solve this problem, and now we have, potentially, this other agent that can be given as an intravitreal injection,” Tarek S. Hassan, MD, a proponent of potential ocriplasmin use, said.

However, at this early stage of treatment with Jetrea (ocriplasmin, ThromboGenics), specialists are not in complete agreement about what should be treated and when, according to Jay S. Duker, MD, OSN Retina/Vitreous Board Member.

First, it is important that physicians establish standard nomenclature for pathology that falls under the heading of vitreomacular interface disease, Duker said, because vitreomacular adhesion (VMA) and vitreomacular traction (VMT) are not synonymous.

“The label for ocriplasmin in the United States is for ‘symptomatic VMA.’ That’s not a terminology that is useful or particularly accurate,” Duker said. “While I like having the broad label so that we physicians have leeway in our decision making, in reality what we are treating is vitreomacular traction.”

There are four approaches to managing vitreoretinal disease: pneumatic vitreolysis, enzymatic vitreolysis with ocriplasmin, vitrectomy surgery and observation.

At the American Society of Retina Specialists meeting in Toronto, Harry W. Flynn Jr., MD, presented results of a study on observation as a viable alternative to other methods of managing vitreomacular disease.

“I’m impressed by how well the patients did without treatment,” Flynn said in a subsequent interview. “In patients having minimal or no symptoms, observation is a reasonable course of management without getting into the invasive and expensive procedures.”

The U.S. Food and Drug Administration approved ocriplasmin for symptomatic vitreomacular adhesion in October 2012.

In September, the U.K.’s National Institute for Health and Care Excellence (NICE) issued a final appraisal determination for ocriplasmin confirming that the drug will be recommended for reimbursement in England and Wales via the National Health Service.

The NICE appraisal committee recommended ocriplasmin as an option for treating vitreomacular traction in adults only if an epiretinal membrane is not present and patients have a stage 2 full-thickness macular hole with a diameter of 400 µm or less and/or severe symptoms.

The committee quoted clinicians who think that ocriplasmin does not “have a place in the treatment of patients with vitreomacular traction and an epiretinal membrane because it was not clinically effective in these patients.”

The committee reported that ocriplasmin was clinically effective and cost-effective in patients with vitreomacular traction and without epiretinal membrane and in patients with vitreomacular traction and macular hole.

MIVI-TRUST trial

The MIVI-TRUST trial, two multicenter, randomized clinical trials conducted by the Microplasmin for Intravitreous Injection-Traction Release without Surgical Treatment (MIVI-TRUST) study group in conjunction with ThromboGenics, was designed to compare a single intravitreal injection of ocriplasmin 125 µg with a placebo injection in patients with symptomatic vitreomacular adhesion.

Results of the MIVI-TRUST trial were published in the New England Journal of Medicine in 2012.

The study included 652 eyes; 464 eyes were treated with ocriplasmin and 188 received the placebo. Vitreomacular adhesion resolved in 26.5% of eyes in the ocriplasmin group and 10.1% of eyes in the placebo group; the difference was statistically significant (P < .001).

Total posterior vitreous detachment was identified in 13.4% of eyes in the ocriplasmin group and 3.7% of eyes in the placebo group (P < .001). Nonsurgical macular hole closure occurred in 40.6% of eyes in the ocriplasmin group and 10.6% of eyes in the placebo group (P < .001).

In addition, eyes in the ocriplasmin group were more likely to gain at least three lines of best corrected visual acuity.

Hassan described the pharmacologic action of ocriplasmin.

“It does two things,” he said. “It cleaves the posterior hyaloid of the vitreous from the retinal surface and the internal limiting membrane of the retina, and it liquefies the vitreous. Its action is as a fibrinolytic enzyme. The medication itself acts on fibronectin and laminin and a number of MMPs, matrix metalloproteinases, and through their actions with those three naturally occurring substances does both of those things.”

Another study showed that early treatment with ocriplasmin may halt disease progression and improve visual outcomes. At the Euretina meeting in September, Peter Stalmans, MD, PhD, reported that 16.4% of eyes diagnosed with vitreomacular adhesion progressed to vitreomacular traction within 1 year. Also, vitreomacular adhesion resolved spontaneously in only 20% of patients with vitreomacular traction, Stalmans said. He said the study was the largest retrospective analysis to date on patients with vitreomacular adhesion and vitreomacular traction.

Classifying vitreomacular disease

A finding of vitreomacular adhesion on optical coherence tomography is normal and does not predict pathology, Duker said.

“Vitreomacular adhesion is part of a natural process of the eye. It’s a part of the natural aging liquefaction process of the vitreous. VMA as an OCT finding is typically normal. So, I don’t treat VMA. I treat vitreomacular traction,” he said.

Duker said there is an overarching need to identify potential pathology in eyes with vitreomacular adhesion.

“The first thing that we need to do is to identify eyes that have VMA that are going to go on to be pathologic,” he said. “We need to figure out the features on OCT or other clinical features that suggest that it’s going to lead to pathology. If we could do that reliably, then we might think about use of ocriplasmin at an earlier stage. But until that time, VMA, in my mind, should not be treated. VMT and full-thickness macular hole are what we are treating with ocriplasmin.”

To better identify eyes with potential pathology, Duker and colleagues developed an international classification system for vitreomacular interface disease. Their paper on the classification system was published online in Ophthalmology in September.

“The problem when you review the literature is that the same entity is called many different things by different authors. We need to speak a common language,” Duker said. “There are definitely eyes that start out with VMA that will develop pathology. But, unfortunately, right now nobody can look at an OCT of an eye with VMA and predict whether or not that eye will go on to VMT, macular hole or epiretinal membrane, or simply have a posterior vitreous detachment and be normal. According to the new classification system, VMA is an OCT finding based on a single B-scan where there is a perifoveal posterior vitreous detachment and the retina is otherwise normal.”

Duker and colleagues established four types of vitreomacular disease: vitreomacular adhesion, full-thickness macular hole, lamellar macular hole and epiretinal membrane.

“Some people would also classify myopic macular schisis as a vitreomacular interface disease. The one thing they all have in common, beside the location of the disease, is the cause: anomalous posterior vitreous detachment. That’s what they share,” Duker said.

Vitreomacular adhesion was subdivided into focal- or broad-based pathology, with focal disease having an adhesion less than 1,500 µm and broad disease being greater than 1,500 µm, Duker said.

“And also whether VMA was isolated or concurrent,” he said. “Isolated means there’s no other disease in the eye. Concurrent means there is another disease such as diabetic macular edema, age-related macular degeneration or retinal vein obstruction. It’s not uncommon to see vitreomacular adhesion in those eyes. We as of yet don’t know if VMA plays a pathologic role with those other diseases and, therefore, if relief of the VMA would be a benefit.”

Duker and colleagues defined full-thickness macular hole as a foveal lesion with interruption of all retinal layers from the internal limiting membrane to the retinal pigment epithelium.

According to Duker, the classification of macular hole involves three factors. The first factor is size of the defect, which can be measured on OCT and correlates with success rates for both vitrectomy and ocriplasmin.

The second factor is the presence or absence of vitreomacular traction.

“That’s also important because if VMT is absent in a macular hole, there’s no reason to consider ocriplasmin therapy,” Duker said.

The third factor is the macular hole being primary or secondary.

“We used to call macular holes idiopathic, which means that we didn’t know the cause,” Duker said. “We do know the cause. ‘Primary’ macular holes are caused by VMT. They start with VMA, go into VMT and then a macular hole occurs. That’s a primary macular hole. Secondary macular holes are caused by something else other than VMA and VMT, for example, blunt trauma or lightning strike. Patients with high myopia can develop full-thickness macular holes without VMT. Those would all be considered secondary macular holes. This classification system is completely OCT-based.”

Hassan said there is a need for a classification system to predict outcomes.

“I think that there’s the potential that such classification systems would be beneficial in streamlining the way that we approach our patients, whether we go to surgery or try ocriplasmin as an initial approach,” Hassan said. “But the value of such a classification system is only seen if it correlates with the outcomes of the procedures performed. So, I’m all in favor of a new classification system if we can correlate it with outcomes.”

Hypothetically, a new classification system may be used to conduct a retrospective analysis of data from the MIVI-TRUST trial, Hassan said.

“Unfortunately, the only data set that really exists in a prospective randomized manner is the MIVI-TRUST trial. So, you’d have to have a new classification system fully in place, then apply it to retrospectively look at the eyes that went through it and see if the classification system would have been ultimately predictive,” Hassan said. “We can take the new system, look at all of the eyes and see how robust it is.”

Closing macular holes

At the ASRS meeting, Peter K. Kaiser, MD, OSN Retina/Vitreous Board Member, presented data from the MIVI-TRUST trial on the effectiveness of ocriplasmin in closing full-thickness macular holes in patients with vitreomacular adhesion.

“We know that in surgery, the smaller the hole is, the higher the success rate,” Kaiser said. “It sort of stands to reason, the same thing would occur with ocriplasmin. It’s just easier to close smaller holes.”

Results showed high rates of closure in eyes with smaller full-thickness macular holes, Kaiser said. Macular hole closure was a secondary outcome in the MIVI-TRUST study.

“This is a treatment that works very well if you choose the right patient,” Kaiser said. “So, if you have patients with vitreomacular traction and small or medium-sized full-thickness macular holes — less than 400 µm in internal dimension, measured within the retina and not at the surface of the retina — those patients do very well with an ocriplasmin injection.”

Among patients treated with ocriplasmin, the rate of closure of full-thickness macular holes was 36.8% in eyes with baseline diameter of 250 µm to 400 µm and 58.3% in eyes with baseline diameter less than 250 µm, Kaiser said.

“This is without surgery, without face-down positioning, without a gas bubble and without cataract surgery,” he said. “So, obviously, very good anatomic results with just an injection. Remember that these patients have to have vitreomacular traction in addition to a full-thickness macular hole. It’s not simply hole size.”

In addition, macular holes closed in some patients who did not experience complete release of vitreomacular traction, Kaiser said.

“The other thing that this study showed was that in some of the patients, the primary outcome of the MIVI-TRUST program wasn’t met. In other words, the traction was not released from the entire macula but the holes still closed because traction was released from the edge of the hole,” he said.

Also, the presence of epiretinal membrane did not seem to affect the rate of successful hole closure, Kaiser said.

“Usually with ocriplasmin, patients who have epiretinal membranes are not good candidates for treatment,” he said. “But in the presence of a macular hole, an epiretinal membrane did not lead to lower closure rates. The presence of an epiretinal membrane and a macular hole doesn’t seem to be a contraindication like it is in patients who have just vitreomacular traction and no macular hole.”

OCT played a key role in choosing patients for treatment and assessing clinical outcomes, Kaiser said.

“The OCT was basically the key determiner of outcomes in the study because patients had to have OCT-based vitreomacular traction. In every patient, there was an OCT performed,” he said. “The primary outcome was also based on OCT. Everything in the study kind of revolved around OCT. It was very important to outcomes, as it is in clinical practice.”

Predictive features of success

While Kaiser explained the effectiveness of ocriplasmin in closing macular holes in the presence of vitreomacular adhesion, Hassan presented MIVI-TRUST trial data highlighting five specific predictive factors for resolution of vitreomacular adhesion and closure of full-thickness macular hole: age younger than 65 years, full-thickness macular hole 400 µm or smaller, vitreomacular adhesion diameter less than 1,500 µm, absence of epiretinal membrane and phakic lens status.

Results at 28 days showed resolution of vitreomacular adhesion in 26.5% of eyes in the ocriplasmin group and 10.1% of eyes in the placebo group, Hassan said. The difference was statistically significant (P < .001).

Among eyes with full-thickness macular holes 400 µm in diameter or smaller, resolution of vitreomacular adhesion was achieved in 50% in the ocriplasmin group and 25.5% in the placebo group. Among patients younger than 65 years, resolution of vitreomacular adhesion was achieved in 47.5% in the ocriplasmin group and 23.3% in the placebo group.

Among eyes with vitreomacular adhesion and no epiretinal membrane, resolution of vitreomacular adhesion occurred in 37.4% in the ocriplasmin group and 14.6% in the placebo group. Among phakic eyes, resolution occurred in 34.2% in the ocriplasmin group and 12.6% in the placebo group.

“All of those percentages were higher than the overall rates of anatomic success from the MIVI-TRUST trial. I think such an elucidation of factors that are predictive of successful pharmacologic vitreolysis is beneficial in the sense that it allows us to counsel our patients as to their likelihood of having a successful outcome more accurately based on their exact presentation, rather than based on results from the whole study population,” Hassan said.

Observation a viable option

Flynn’s study on observation included 106 eyes of 81 patients with vitreomacular adhesion identified on OCT. The individual treating physician selected observation as the management strategy in this study. There were no patients with macular hole at study entry.

“This was a study that was intended to look at the natural history of this problem with respect to how often they progress and how often bad things happen like macular hole or progressive loss of vision. Similarly, it looked at good things that might happen, like spontaneous release of vitreomacular adhesion,” Flynn said. “Our project was intended to back up beyond the treatment approaches and look back at observation. Our second purpose was to propose a classification scheme for vitreomacular adhesion based on spectral-domain OCT findings.”

Flynn and colleagues designated three grades of vitreomacular adhesion based on OCT images. Grade 1 was incomplete separation of the cortical vitreous with foveal attachment. Grade 2 was the criteria of grade 1, plus an intraretinal cyst. Grade 3 was criteria of the first two grades, plus the presence of subretinal fluid under the fovea.

“We use the term ‘grades’ because we don’t know that the natural course follows a linear stage of progression. So, if we had called it a stage, that would imply that one stage leads to the next stage,” Flynn said. “We don’t really know that. We use the term grade because we simply look at the OCT and we grade it. Then we can categorize patients and compare the baseline pictures with follow-up pictures.”

Forty-three eyes presented with grade 1 vitreomacular adhesion, 56 eyes presented with grade 2, and seven eyes presented with grade 3, Flynn said.

Baseline Snellen visual acuity was 20/32 in grade 1 eyes, 20/40 in grade 2 eyes and 20/54 in grade 3 eyes.

At final follow-up, visual acuity was 20/32 in grade 1 eyes, 20/42 in grade 2 eyes and 20/44 in grade 3 eyes.

Overall, spontaneous release of vitreomacular adhesion occurred in 34 eyes (32%). Spontaneous release occurred in 13 eyes (30%) in the grade 1 group, 17 eyes (30%) in the grade 2 group and four eyes (57%) in the grade 3 group.

Among grade 1 eyes, 23 (53%) remained stable, with no release of vitreomacular adhesion, and seven (16%) worsened. Among grade 2 eyes, 31 (55%) remained stable and eight (14%) worsened. Among grade 3 eyes, one (14%) remained stable and two (28%) worsened.

In this study with a mean follow-up of 23 months, five eyes underwent vitrectomy, Flynn said.

“The clinical course of patients with vitreomacular adhesion managed by initial observation is generally favorable in the sense that they had stable vision between the initial and last examinations at a mean of 23 months follow-up,” he said.

The success of observation hinges on symptoms, Kaiser said.

“It really depends on symptoms because you have a very good treatment in the form of surgery and you have a very good treatment in the form of Jetrea, but neither one of those treatments is foolproof and both have their own side effects, pluses and minuses. So, it really comes down to how symptomatic the patient is, whether they would be a good candidate for treatment or not,” he said. – by Matt Hasson