Results of a 2-year phase IIIb study include findings consistent with earlier information about treatment success with ocriplasmin injection for symptomatic vitreomacular adhesion and new information about efficacy and safety over the longer term.
By Cheryl Guttman Krader; Reviewed by Pravin U. Dugel, MD
Findings from a recent randomized trial-the Ocriplasmin (Jetrea, ThromboGenics) for Treatment for Symptomatic Vitreomacular Adhesion Including Macular Hole (OASIS)-provide important clues about the vitreolytic agent’s longer-term safety and efficacy and corroborate the value of careful patient selection for improving treatment success, said Pravin U. Dugel, MD.
“The results from OASIS give us greater confidence about the MIVI-TRUST pivotal trial data,” said Dr. Dugel, managing partner, Retinal Consultants of Arizona, Phoenix, and clinical professor, USC Roski Eye Institute, Keck School of Medicine, University of Southern California, Los Angeles.
“We learned from OASIS that when treatment with ocriplasmin succeeds, the patients continue to do well for at least 2 years,” Dr. Dugel said.
OASIS showed that patients who do not respond to ocriplasmin are not being handicapped in terms of their likelihood to achieve a good visual outcome after vitrectomy, Dr. Dugel said.
The results of OASIS also reinforced the findings from the post-hoc analyses of the pivotal MIVI-TRUST trials and of subsequently reported case studies from numerous centers showing that with better patient selection, the success rate with ocriplasmin treatment goes up, he said.
In addition, OASIS mitigated concerns about early electroretinogram (ERG) changes after ocriplasmin injection and no new safety signals emerged, he noted.
OASIS was a phase IIIb, double-masked clinical trial conducted to further evaluate the long-term efficacy and safety of a single intravitreal injection of ocriplasmin 0.125 mg in patients with symptomatic vitreomacular adhesion (VMA, also known as vitreomacular traction), including full-thickness macular hole (FTMH).
Patients were eligible if they had VMA and best-corrected visual acuity (BCVA) of 20/32 or worse in the study eye. Key exclusion criteria (assessed by the study investigator) were presence of aphakia, FTMH >400 μm, and epiretinal membrane (ERM) in the study eye.
“It is noteworthy that patients in OASIS were evaluated with spectral-domain optical coherence tomography (OCT), which is the standard of care for assessing the retina in eyes with VMA and FTMH, whereas only time-domain OCT was available when MIVI-TRUST was conducted,” Dr. Dugel said. “Use of the advanced OCT technology is valuable for informing us about the long-term efficacy and safety profile of ocriplasmin.”
A total of 220 patients were randomly assigned 2:1 to ocriplasmin or sham. The primary endpoint-proportion of patients with pharmacologic VMA resolution-was assessed at day 28.
“In the pivotal MIVI-TRUST trials, ocriplasmin was compared with injection with vehicle, which may have induced VMA resolution through a mechanical or volume effect,” Dr. Dugel said. “Use of sham injection as the comparator in OASIS allowed a clearer assessment of the true treatment effect of ocriplasmin.”
The ocriplasmin and sham study groups were similar in their demographics and baseline ocular characteristics.
Assessment of baseline OCT images at the central reading center, however, showed that a sizable number of patients in each group should have been excluded from study participation per the eligibility criteria. VMA was absent in 1.4% of sham patients and 4.1% of ocriplasmin patients, about 23% of patients in each group had an ERM, and 15.4% of sham patients and 20% of ocriplasmin patients had a FTMH >400 μm in diameter
OASIS met its primary endpoint as the proportion of patients achieving VMA release at day 28 was significantly higher in the ocriplasmin group than in the controls (41.7% versus 6.2%; p < 0.001). The treatment benefit was maintained through ongoing follow-up to the end of the study.
“The success of ocriplasmin in OASIS compares favorably with the results of the MIVI-TRUST trials in which it was associated with a VMA resolution rate was 26.5%,” Dr. Dugel said. “Not only was the resolution rate higher with ocriplasmin in OASIS, which reflects better patient selection, but OASIS also provides data from 2 years of follow-up rather than just 6 months.”
Further reinforcing the importance of patient selection, results of subgroup analyses showed that ocriplasmin-treated patients who met all eligibility criteria had higher VMA resolution rates than their counterparts with protocol violations.
In addition, the difference in resolution rates between the ocriplasmin and sham groups was greater among eyes that had no protocol violations than in those with protocol violations and in eyes with baseline ocular characteristics previously shown to predict success (e.g., focal adhesion ≤1,500 μm, FTMH) than in those without those prognostic findings.
For example, the VMA resolution rate at day 28 after ocriplasmin treatment was 45.4% in patients with a focal adhesion (≤1,500 μm) versus 16.7% in eyes with a broader attachment and 50.6% in eyes without an ERM but only 13.1% if an ERM was present. Rates of release in the sham group where 7.3% in eyes with a focal adhesion, 0% in those with a broader adhesion, 8.4% if an ERM was absent, and 0% if an ERM was present.
A secondary efficacy endpoint in OASIS analyzed the proportion of patients achieving BCVA improvement ≥2 lines from baseline at month 24, irrespective of vitrectomy. Its results showed no statistically significant difference between the ocriplasmin and sham treatment groups (50.5% versus 39.1%; p = 0.114).
“This finding gives me confidence about waiting a month to perform vitrectomy in a patient who is appropriate for ocriplasmin,” Dr. Dugel said. “In case the treatment is not successful, however, when patients choose pharmacologic intervention rather than primary vitrectomy, I still set them up to be on the surgery schedule after about 1 month. If ocriplasmin is successful, the surgery can easily be canceled. Having a date set up in advance makes it easier for the patient and for surgery scheduling.”
Adverse events associated with ocriplasmin treatment in OASIS were consistent with those previously reported with respect to type and incidence.
Data on adverse events of special interest showed new onset or progression of macular hole occurred in 15.8% of ocriplasmin eyes and 13.5% of the sham group. Rates of visual acuity reduction were 12.3% in the ocriplasmin group and 24.3% for sham. Subretinal fluid developed in 11.6% of ocriplasmin eyes and 4.1% of the controls.
OASIS also featured an ERG substudy, which included 40 ocriplasmin-treated patients and 21 patients in the sham group. The rate of expert defined ERG abnormalities (reduction ≥40%) starting on study day 7 or 28 in the study eye was about eight-fold higher in the ocriplasmin group compared with sham, 40% (16/40) versus 4.8% (1/21).
However, 13 of the ERG reductions in the ocriplasmin group and the single case in the sham group resolved after a median of about 6 months, and 94% of eyes with an ERG abnormality maintained or gained BCVA.
“As a matter of fact, there was a tendency for patients who developed ERG changes to have better anatomic and functional outcomes than their counterparts who had no ERG abnormalities,” Dr. Dugel said. “These findings, however, should not lead retina specialists to ignore the potential for ERG changes to occur after ocriplasmin treatment.
“This is an important safety issue that needs to be discussed with patients and that requires follow-up,” he added. “These and other findings from OASIS show us that we still have a lot to learn about how ocriplasmin works and its physiologic effects.”
Pravin U. Dugel, MD
This article was adapted from Dr. Dugel’s presentation at the American Academy of Ophthalmology meeting and the published report of the OASIS trial results [Dugel PU, et al. Ophthalmology. 2016;123:2232-2247]. Dr. Dugel is a consultant to Alcon Laboratories, Novartis, and ThromboGenics.