Vitreous Levels of Active Ocriplasmin Following Intravitreal Injection: Results of an Ascending Exposure Trial

Abstract

Purpose.: To characterize the levels of active ocriplasmin over a period of ascending time points (range, 5 minutes to 7 days) from intravitreal injection of a 125-μg dose to sampling.

Methods.: During this 7-week controlled, open-label, phase 2 study, a single intravitreal (125 μg) dose of ocriplasmin was injected into the midvitreous of one eye of each of 34 patients prior to their scheduled primary pars plana vitrectomy. Patients were allocated to vitreous sampling at the beginning of the surgery, which occurred 5 to 30 minutes, 31 to 60 minutes, 2 to 4 hours, 24 ± 2 hours, or 7 ± 1 days after ocriplasmin injection, or to the control group, who received no ocriplasmin injection.

Results.: With increasing time from ocriplasmin injection to vitreous sampling, mean active ocriplasmin concentration decreased. While at 5 to 30 minutes postinjection, mean active ocriplasmin concentration was 11,597.7 ng/mL, within 31 to 60 minutes from injection the mean active ocriplasmin concentration had reduced to 8108.7 ng/mL; and by 24 hours after injection, half of the patients (2/4) had active ocriplasmin concentrations below the lower limit of quantification (LLQ; <272.4 ng/mL), as did all samples from the day 7 and control groups. No ocular serious adverse events (SAEs) were reported in patients who received ocriplasmin, while three ocular SAEs occurred in the study eye of one patient in the control group (1/38; 2.6%).

Conclusions.: Active ocriplasmin concentrations in vitreous samples decreased with increasing time from injection to sample, with enzyme levels in all of the patients in the day 7 group being comparable to those in the control group. (ClinicalTrials.gov number, NCT01159665.)

Introduction

The vitreoretinal interface of the human eye is a complex anatomical structure where the vitreous gel attaches to the inner limiting membrane of the retina via collagen fibrils and adhesive molecules including fibronectin and laminin. ^{1 –3} Over time, the vitreous gel undergoes liquefaction accompanied by progressive weakening of the adhesion at the vitreoretinal interface. These processes lead to a separation of the vitreous from the retina, known as posterior vitreous detachment (PVD), the early signs of which are evident from 40 years of age onward. ^{4 –7}  

Persistent vitreomacular adhesion (VMA) results when only a partial PVD occurs and part of the posterior hyaloid remains attached to the macular area. ⁸ This persistent adhesion can lead to traction at the vitreoretinal interface, often referred to as vitreomacular traction (VMT), which in turn can lead to symptoms including metamorphopsia, decreased visual acuity (VA), and central visual field defects. ⁹ Based on the literature, there is a known association between VMT (or symptomatic VMA) and diabetic retinopathy, ^10,11 and similarly with exudative age-related macular degeneration. ¹²  

Vision in patients with untreated VMT may deteriorate over time, while spontaneous vitreomacular separation occurs infrequently. ⁹ In a study of patients with VMT who underwent vitrectomy, greater improvement in VA was observed in eyes with shorter duration of symptoms. ¹³ These data suggest that earlier treatment of VMT may help achieve a better visual outcome. 

Progression of VMT can lead to full-thickness macular hole (FTMH) formation, which if left untreated can lead to central legal blindness. ¹⁴ Based on data from observational and randomized trials, conducted before surgery became the standard of care for treatment of FTMHs, the spontaneous closure of FTMHs is a rare occurrence. ^14,15 Rather, the majority (>70%) of small macular holes (MHs) tend to enlarge in diameter, and VA decreases. ¹⁶ Furthermore, with longer duration of FTMHs, there is a greater likelihood of irreversible retinal damage and associated lack of functional improvement following surgery. ¹⁷  

Until recently, the only treatment for VMT, MH, and other diseases related to abnormalities of the vitreoretinal interface was vitrectomy. ¹⁸ Vitrectomy has costs and risks associated with it, and the risks can limit safety and efficacy. ¹ As a result, most eyes with VMT have been observed, rather than treated, until significant loss of vision has already occurred. ¹⁹ A minimally invasive release of vitreous traction could potentially be paradigm shifting in the management of these patients. 

Ocriplasmin (generic name for microplasmin) is a truncated form of the serine protease plasmin that has retained its enzymatic properties. ^20–22 Recently approved for use in the United States and Europe, ocriplasmin is a potent collagenase activator with proteolytic activity against a range of components of the vitreoretinal interface, including fibronectin and laminin. ²³ In vitro, ocriplasmin has the highest activity against fibrinogen, followed by fibronectin, gelatin, collagen type IV, and laminin (ThromboGenics, unpublished data, 2011). The proteolytic activity of ocriplasmin causes disruption of the vitreoretinal interface while preserving the morphology of the retina and the ultrastructure of the inner limiting membrane. ^21,22 Previous preclinical and clinical studies have highlighted the beneficial effects of ocriplasmin in facilitating PVD. Experiments carried out in vivo and ex vivo demonstrated that intravitreal injection of ocriplasmin produced pharmacologic vitreolysis with detachment of the vitreous from the retinal surface. ^{20 –22} Phase 2 studies demonstrated that ocriplasmin was well tolerated and produced nonsurgical resolution of VMA and closure of FTMHs. ^1,11 Results of two phase 3 clinical trials showed a statistically significant difference in favor of a single intravitreal injection of ocriplasmin 125 μg versus placebo injection for achieving the primary efficacy end point, VMA resolution at day 28, as well as the key secondary efficacy end point, total PVD at day 28. ²⁴ Compared with placebo injection, eyes treated with ocriplasmin were more likely to have nonsurgical closure of FTMHs at day 28 (P < 0.001), nonsurgical three-line or greater improvement in best-corrected visual acuity (BCVA) at 6 months (P = 0.02), and a reduced need for vitrectomy at 6 months after ocriplasmin treatment (P = 0.02) (secondary end points). ²⁴  

Additionally, the inactivation profile of ocriplasmin in human vitreous fluid, obtained from random vitrectomy patients, has been investigated over time. A study by de Smet et al. ²⁵ concluded that after 72 hours of incubation at 37°C in human pooled vitreous fluid, less than 0.6% of initial active ocriplasmin was detected. In nonclinical studies using homogenized porcine vitreous, minimal active ocriplasmin was detected after 24 hours, with the inactivation of ocriplasmin following a second-order process. ²⁶  

Herein, we report the results of a single-center study to evaluate the vitreous levels of active ocriplasmin (125-μg dose) when injected intravitreally at ascending time points prior to planned primary vitrectomy. The safety of ocriplasmin was also analyzed through consideration of adverse events (AEs) and postinjection complications. 

Methods

Study Design and Population

This 7-week, single-center, open-label, phase 2 study was designed to evaluate the vitreous levels of active ocriplasmin following a single intravitreally injected 125-μg dose of ocriplasmin in patients scheduled for primary vitrectomy. The planned sample size was 36 patients in whom a vitreous sample was obtained at the beginning of vitrectomy for determination of ocriplasmin activity. Patients were allocated to one of six groups in a sequential manner until a group was full, starting with group 1 and ending with group 6: 

•  
  Group 1: vitreous sampling 5 to 30 minutes after ocriplasmin injection (n = 8);
•  
  Group 2: vitreous sampling 31 to 60 minutes after ocriplasmin injection (n = 8);
•  
  Group 3: vitreous sampling 2 to 4 hours after ocriplasmin injection (n = 8);
•  
  Group 4: vitreous sampling 24 ± 2 hours after ocriplasmin injection (n = 4);
•  
  Group 5: vitreous sampling 7 ± 1 days after ocriplasmin injection (n = 4); and
•  
  Group 6: vitreous sampling without prior ocriplasmin injection (control arm) (n = 4).

The time points for vitreous sampling were selected based on preclinical data. ²⁵ The aim was to include early time points when ocriplasmin inactivation was expected to be the fastest, as well as later time points to identify a time when ocriplasmin was no longer active in all patients. 

While the plan was to enroll 36 patients, a total of 38 were enrolled, with one extra patient each allocated to group 1 and group 2 (see Patient Demographics and Disposition). 

Patients were aged ≥18 years with eye disease for which primary vitrectomy was indicated, although certain types of eye disease, previous treatments, or pregnancy were criteria for exclusion (Table 1). 

Table 1. 

 

Key Eligibility and Exclusion Criteria

Inclusion criteria

Male or female patients aged ≥18 y

Eye disease for which primary vitrectomy was scheduled

Best-corrected visual acuity of 20/800 or better in the nonstudy  eye

Exclusion criteria

Proliferative diabetic retinopathy

Patients with any vitreous hemorrhage or any other vitreous opacification that precluded either of the following: visualization of the posterior pole by visual inspection or adequate assessment of the macula by optical coherence tomography and/or fluorescein angiogram in the study eye

Aphakia in the study eye

High myopia (more than 8 diopters) in study eye (unless prior cataract extraction or refractive surgery made refraction assessment unreliable for myopia severity approximation, in which case axial length >28 mm was an exclusion)

Patients with history of rhegmatogenous retinal detachment in either eye

Patients who had ocular surgery, laser photocoagulation treatment, or intravitreal injection(s) in the study eye in the prior 3 mo

Patients who had laser photocoagulation to the macula of the study eye at any time

Patients with uncontrolled glaucoma in the study eye (defined as intraocular pressure ≥26 mm Hg in spite of treatment with antiglaucoma medication)

Patients with a history of uveitis in either eye

Patients who had previously participated in this study

In groups 1 through 5, ocriplasmin (obtained from ThromboGenics NV, Leuven, Belgium) was administered in the midvitreous of one eye per patient by a single 0.1-mL injection containing 125 μg ocriplasmin, followed by a study eye examination to exclude retinal nonperfusion or other complications. Following the passage of the appropriate allotted time from injection to vitreous sampling, at the beginning of vitrectomy an undiluted 0.5-mL vitreous sample was obtained for analysis of active ocriplasmin levels using a 23-gauge vitrectome (DORC/Dutch Ophthalmic, Zuidland, The Netherlands). Obtaining the sample did not alter the standard surgical approach to vitreous removal during vitrectomy. The elapsed time from the beginning of the vitrectomy until achievement of complete core vitrectomy was calculated. 

Evaluations were performed at baseline, on the day of the surgery, and postsurgery over a period of 7 weeks at study visits on days 1, 14 (±2 days), and 42 (±3 days). Ophthalmic evaluations measured intraocular pressure (IOP) and BCVA, and included slit-lamp examinations, optical coherence tomography (OCT), and dilated ophthalmoscopy. 

The study protocol and informed consent forms were reviewed and approved by an Ethics Committee at the investigative site and the Belgian Competent Authority. The study was conducted in compliance with the study protocol, the Principles of International Conference on Harmonisation Good Clinical Practice, and all federal and local requirements. The study was monitored and managed by a contract research organization, Chiltern International Limited. The study adhered to the tenets of the Declaration of Helsinki. Following explanation of the nature and possible consequences of the study, patients signed an informed consent form before entering the trial. 

Analysis of the Concentration of Active Ocriplasmin

The determination of the concentration of active ocriplasmin in the vitreous samples involved measurement of its enzymatic activity toward a chromogenic substrate, S-2403. The samples were diluted if necessary and incubated at 37°C with substrate, and then chromogen formation was measured by spectrophotometer-detected increases in absorbance per second at 405 nm. The formation of a chromogen product is proportional to the enzymatic activity of ocriplasmin, so the concentration of active ocriplasmin in the vitreous fluid could be calculated using calibration curve parameters. 

Calibration Curves.

Calibration curves were prepared once daily from 1000 nM and 200 nM stock solutions of ocriplasmin; each calibration curve was based on the readings from three 96-well plates. The stock solutions were prepared at 4°C by diluting ocriplasmin in NaCl – 5 mM citrate buffer. The ocriplasmin stock solutions were then diluted in test buffer, which consisted of human vitreous fluid diluted with an equal volume of NaCl – 100 mM citrate buffer, to prepare the standard curve solutions (100, 70, 40, 20, 10, and 5 nM). 

Tris buffer 200 mM was heated to 37°C, added to 96-well plates (240 μL/well), and incubated for at least 5 minutes at 37°C. Each standard curve solution was added into 4 wells (30 μL/well) and incubated for at least 2 minutes at 37°C before addition of the S-2403 substrate (30 μL 3 mM solution). The solutions were mixed for at least 10 seconds using a Thermomixer (Eppendorf, Hamburg, Germany). The absorbance at 405 nm was read every 20 seconds for 5 minutes in a microplate spectrophotometer. The mean change in absorbance per minute was calculated between 1 and 5 minutes for each ocriplasmin solution (100, 70, 40, 20, 10, and 5 nM), and a standard curve was prepared. 

Determination of Active Ocriplasmin Concentration.

Study vitreous samples (1 mL) were thawed at 2°C to 8°C and homogenized by passing twice through a 5-mL syringe and then three to five times through a 5-mL syringe capped with an 18-gauge 1.5-inch needle until the matrix was broken. The samples were thoroughly mixed and then centrifuged for 5 minutes at 12,000g at 2°C to 8°C. The supernatant was stored on ice and diluted in test buffer, if required. 

Tris buffer 200 mM was heated to 37°C, added to 96-well plates (240 μL/well), and incubated for at least 5 minutes at 37°C. Each quality control sample (three or four samples per plate of 80, 50, 20, and/or 10 nM; two preparations per quality control level) or study vitreous sample was added into 4 wells (30 μL/well) and incubated for at least 2 minutes at 37°C before addition of the S-2403 substrate (30 μL 3 mM solution). The solutions were mixed, and absorbance was read as described above (see Calibration Curves). The mean change in absorbance per minute was calculated between 1 and 5 minutes for each sample, and the concentration of active ocriplasmin was calculated from the calibration curve. The calibration curve was accepted if the back-calculated concentration of calibration standards was within ±30% of the nominal concentration. 

Data Analysis

The analysis population consisted of all patients who received one dose of study medication or were part of the control arm (group 6). Summary statistics were calculated for continuous variables describing the number of observations (n), mean, standard deviation (SD), median, minimum, and maximum. Group frequencies and percentages were presented for categorical variables. Percentages were calculated using the total number of patients per treatment group or number of patients with nonmissing data, as appropriate. 

Assessment of Active Ocriplasmin Levels.

The primary end point of this study was the concentration of active ocriplasmin in vitreous samples obtained at the beginning of vitrectomy in patients at various time points relative to ocriplasmin injection (postinjections) as follows: group 1, 5 to 30 minutes; group 2, 31 to 60 minutes; group 3, 2 to 4 hours; group 4, 24 ± 2 hours; group 5, 7 ± 1 days. For each treatment group, the concentration of active ocriplasmin was summarized by descriptive statistics, and then a plot was generated with the group means (SEM) of both the concentrations of active ocriplasmin and group sampling times with a semilogarithmic (log10) transformation of the x-axis. Individual data were also plotted as the concentration of active ocriplasmin decreased over time after semilogarithmic (log10) transformation of the x-axis (time). 

The second-order rate inactivation constant of ocriplasmin was determined based on a categorized data analysis. Data were categorized to time before fitting in GraphPad Prism v5 (GraphPad Software, La Jolla, CA) with the second-order regression formula Y = 1/((1/A ₀) + k × X) where Y stands for molar concentration, A ₀ is the starting molar concentration, X represents time in seconds, and k is the second-order constant. Initial fitting constraints required both k and A ₀ to be greater than 0. 

Safety Assessments.

The safety end points of this study were a summary of postinjection AEs. Safety summaries included only treatment-emergent AEs, defined as events with an onset on or after the date of study drug injections. For the control arm (group 6), AEs on or after the baseline visit were considered treatment emergent to allow comparison with the treatment groups. All AEs were coded using the Medical Dictionary for Regulatory Activities (MedDRA), Version 13.1. All AEs and serious AEs (SAEs) were summarized by treatment group, system organ class, and preferred term. If a patient had multiple occurrences of an AE, the patient was included only once in the respective patient count for that AE. 

Results

Patient Demographics and Disposition

A total of 38 patients were enrolled in the study, two more than the planned size of 36; two additional patients were enrolled after vitreous samples of two patients were excluded from the analysis of time-dependent ocriplasmin activity. One patient had a previous vitrectomy and retinal detachment in the study eye (violation of exclusion criteria), while the vitreous sample from the other patient was contaminated with an irrigating solution used during vitrectomy. All 38 patients were included in the safety analysis, while 36 were included in the analysis of ocriplasmin active enzyme concentration. Therefore, the study group sizes were as follows: 

•  
  Group 1: vitreous sampling 5 to 30 minutes after ocriplasmin injection (n = 9);
•  
  Group 2: vitreous sampling 31 to 60 minutes after ocriplasmin injection (n = 9);
•  
  Group 3: vitreous sampling 2 to 4 hours after ocriplasmin injection (n = 8);
•  
  Group 4: vitreous sampling 24 ± 2 hours after ocriplasmin injection (n = 4);
•  
  Group 5: vitreous sampling 7 ± 1 days after ocriplasmin injection (n = 4); and
•  
  Group 6: vitreous sampling without prior ocriplasmin injection (control arm) (n = 4).

The study population demographics at baseline are described in Table 2; 100% of the population were white, 32.4% were male in the ocriplasmin groups, and 25% were male in the control group. Within the first five groups, the underlying reasons for vitrectomy were vitreous floaters (32.4%), FTMH (26.5%), macular pucker (17.6%), VMT (5.9%), and other (17.6%). The control group underwent vitrectomy because of diabetic retinopathy (50.0%), FTMH (25.0%), and macular pucker (25.0%). Of the enrolled patients, 37 of 38 (97.4%) completed the study. One patient allocated to group 2 was discontinued from the study due to withdrawal of consent, as the patient did not want to attend the follow-up visits. 

Table 2. 

 

Patient Demographics at Baseline

	Ocriplasmin						Control
	Group 1, n = 9	Group 2, n = 9	Group 3, n = 8	Group 4, n = 4	Group 5, n = 4	Total Active, n = 34	Group 6, n = 4
Male, n (%)	2 (22.2)	3 (33.3)	2 (25.0)	1 (25.0)	3 (75.0)	11 (32.4)	1 (25.0)
Mean age, y (SD)	71.3 (8.56)	65.6 (13.50)	63.8 (9.38)	62.5 (8.66)	59.3 (11.87)	65.6 (10.77)	69.8 (7.80)
Reason for vitrectomy, n (%)
Macular pucker	1 (11.1)	2 (22.2)	2 (25.0)	1 (25.0)	0	6 (17.6)	1 (25.0)
Macular hole	4 (44.4)	0	4 (50.0)	1 (25.0)	0	9 (26.5)	1 (25.0)
Floaters	1 (11.1)	3 (33.3)	1 (12.5)	2 (50.0)	4 (100.0)	11 (32.4)	0
Vitreomacular traction	1 (11.1)	0	1 (12.5)	0	0	2 (5.9)	0
Diabetic retinopathy	0	0	0	0	0	0	2 (50.0)
Epiretinal membrane	0	0	0	0	0	0	0
Other	2 (22.2)	4 (44.4)	0	0	0	6 (17.6)	0
IOL luxation	1 (11.1)	1 (11.1)	0	0	0	2 (5.9)	0
Lens luxation	0	2 (22.2)	0	0	0	2 (5.9)	0
Macular hemorrhage	0	1 (11.1)	0	0	0	1 (2.9)	0
Opacification of IOL	1 (11.1)	0	0	0	0	1 (2.9)	0

 IOL, intraocular lens.

Assessment of Active Ocriplasmin Levels

Vitreous samples were collected at ascending time points from the injection of ocriplasmin. In vitreous samples collected at 5 to 30 minutes postinjection (group 1), the mean active ocriplasmin concentration was 11,597.7 ng/mL. With increasing time from ocriplasmin injection to vitreous sampling, the mean active ocriplasmin concentration decreased (Figure). Within 31 to 60 minutes from injection, the mean concentration was 8108.7 ng/mL (group 2); within 2 to 4 hours, it was 2610.6 ng/mL (group 3); and within 24 hours it was 496.5 ng/mL (group 4). Half of the patients (2/4) in group 4 had active ocriplasmin concentrations below the lower limit of quantification (LLQ; <272.4 ng/mL). Seven days after the injection of ocriplasmin, the mean active enzyme concentrations in the vitreous samples (group 5) were comparable to those of the control group given no ocriplasmin injection (group 6). All of the active enzyme concentrations in groups 5 and 6 were below the LLQ. The second-order rate inactivation constant of ocriplasmin was 925.6 ± 427.5 (SE) M⁻¹ s⁻¹ (95% confidence interval [CI]: 52.61–1799). 

Figure. 

 

Mean active ocriplasmin concentrations (±SEM) decreased with increasing time from midvitreous injection to sampling.

The duration of each vitrectomy was calculated as the time the vitrectomy was initiated to the time of the end of the core vitrectomy phase. The mean duration of vitrectomy ranged between 3.7 minutes (group 2) and 5.1 minutes (group 3), with the individual range between 1 and 12 minutes. Overall, there was no notable difference in the duration of vitrectomy between treatment groups. 

Safety

Overall, a single intravitreal injection of ocriplasmin 125 μg was well tolerated by patients when administered at different time points prior to planned primary vitrectomy. There were 76 treatment-emergent AEs, including 65 AEs in 21 of 34 patients (61.8%) who received ocriplasmin (groups 1–5) and 11 AEs in 3 of 4 patients (75.0%) who did not receive ocriplasmin (group 6; control group). The majority of AEs were mild in intensity and resolved with no sequelae, as did the majority of the ocular AEs. Furthermore, the majority of AEs were ocular (67/76), mild in intensity (62/76), and in the study eye—with the exception of two AEs that occurred in the nonstudy eye. Drug-related AEs occurred in 8 of 34 patients (23.5%), all of which were localized to the study eye. 

Overall, 4 of 38 patients (10.5%) reported at least one treatment-emergent SAE (six events in total), three of which were ocular events (Table 3). Two of the SAEs occurred in 2 of 34 patients (5.9%) who had received ocriplasmin, and four SAEs occurred in 2 of 4 patients (50.0%) in the control group (Table 3). No ocular SAEs were reported in patients who received ocriplasmin, while three ocular SAEs occurred in the study eye of one patient in the control group (1/38; 2.6%, Table 3). No deaths occurred during the study, and no patients were discontinued because of AEs or SAEs. 

Table 3. 

 

Treatment-Emergent Serious Adverse Events

Treatment-Emergent SAEs	Ocriplasmin		Control
	n (%); Total = 34	Events	n (%); Total = 4	Events
Eye disorders	0 (0)	0	1 (25.0)	3
Choroidal hemorrhage	0 (0)	0	1 (25.0)	1
Hyphema	0 (0)	0	1 (25.0)	1
Vitreous hemorrhage	0 (0)	0	1 (25.0)	1
Cardiac disorders	0 (0)	0	1 (25.0)	1
Congestive cardiac failure	0 (0)	0	1 (25.0)	1
Neoplasms benign, malignant, and unspecified (including cysts and polyps)	1 (2.9)	1	0 (0)	0
Paraneoplastic syndrome	1 (2.9)*	1*	0 (0)	0
Respiratory, thoracic, and mediastinal disorders	1 (2.9)	1	0 (0)	0
Respiratory failure	1 (2.9)*	1*	0 (0)	0

*  Adverse events considered unrelated to study drug.

Ocular AEs.

Ocular AEs among the 34 patients who received ocriplasmin are presented in Table 4. Among the groups who received ocriplasmin (groups 1–5), 19 of 34 patients (55.9%) had a total of 59 ocular AEs, of which 57 occurred in the study eye. In the study eyes of patients treated with ocriplasmin, 29 of 57 AEs occurred previtrectomy and 28 of 57 occurred postvitrectomy. Previtrectomy ocular AEs increased as the time increased from ocriplasmin injection to vitreous sampling; the majority of AEs were reported in groups 4 and 5, where vitreous sampling occurred 24 ± 2 hours and 7 ± 1 days after ocriplasmin injection, respectively. In the control group, 2 of 4 patients (50.0%) experienced a total of eight ocular AEs, all of which were in the study eye. Pre- and postvitrectomy ocular AEs are summarized in Table 4. 

Table 4. 

 

Ocular Adverse Events Pre- and Postvitrectomy

AE Study Eye	Ocriplasmin										Control		Ocriplasmin
	Group 1, n = 9		Group 2, n = 9		Group 3, n = 8		Group 4, n = 4		Group 5, n = 4		Group 6, n = 4		Total Ocriplasmin, n = 34
Patients with any ocular AE, n (%)	2 (22.2)		4 (44.4)		5 (62.5)		4 (100)		4 (100)		2 (50.0)		19 (55.9)
Any study eye event, n	4		5		9		15		24		8		57
Pre- or postvitrectomy	Pre	Post	Pre	Post	Pre	Post	Pre	Post	Pre	Post	Pre	Post	Pre	Post
Blepharitis	0	1	0	0	0	1*	0	0	0	0	0	0	0	2†
Choroidal hemorrhage	0	0	0	0	0	0	0	0	0	0	0	1	0	0
Chromatopsia	0	0	0	0	0	0	2	0	2	0	0	0	4	0
Conjunctival hemorrhage	0	0	0	0	0	0	0	0	1	0	0	0	1	0
Corneal erosion	0	0	0	0	0	0	0	0	0	0	0	1	0	0
Cortical cataract	0	0	0	0	0	0	0	0	0	1	0	0	0	1
Dry eye	0	0	0	0	0	1	0	0	0	0	0	0	0	1
Eye infection	0	0	0	0	0	0	0	2	0	0	0	0	0	2
Eye irritation	0	1	0	0	0	1	0	0	0	0	0	0	0	2
Eye pain	0	0	0	1	0	0	1	0	0	0	0	0	1	1
Hyphema	0	0	0	0	0	0	0	0	0	0	0	1	0	0
Hypotony of the eye	0	0	0	0	0	0	0	0	0	1	0	0	0	1
Increased intraocular pressure	0	2	0	3	0	2	0	3	0	3	0	2	0	13
IOL dislocation	0	0	0	0	0	2	0	0	0	0	0	0	0	2
Macular edema	0	0	0	1	0	2	0	0	0	0	0	0	0	3
Metamorphopsia	0	0	0	0	0	0	0	0	1	0	0	0	1	0
Photophobia	0	0	0	0	0	0	0	0	1	0	0	0	1	0
Photopsia	0	0	0	0	0	0	1	0	4	0	0	0	5	0
Pupillary deformity	0	0	0	0	0	0	0	0	1	0	0	0	1	0
Retinal artery occlusion	0	0	0	0	0	0	0	0	0	0	0	1	0	0
Subretinal fluid	0	0	0	0	0	0	0	0	1	0	0	0	1	0
Retinal tear	0	0	0	0	0	0	0	0	0	0	0	1	0	0
Visual acuity reduced	0	0	0	0	0	0	3	0	4	0	0	0	7	0
Vitreous floaters	0	0	0	0	0	0	3	0	4	0	0	0	7	0
Vitreous hemorrhage	0	0	0	0	0	0	0	0	0	0	0	1	0	0

n  : If a patient had multiple occurrences of an AE, the patient is presented only once in the respective patient count. Events: AE counted each time per patient (n).

*  This event occurred in the nonstudy eye; all other events were in the study eye.

†  This total included one event in the study eye and one in the nonstudy eye.

Postoperative Evaluations.

Slit-lamp, IOP, dilated fundus, and full retinal examinations did not identify any unanticipated safety findings. Dilated retinal examinations did not reveal any vitreous hemorrhage, retinal tear, retinal detachment, or evidence of epiretinal fibrocellular proliferation. While OCT examinations at baseline indicated evidence of FTMH in 10 of 38 patients (26.3%) and epiretinal membrane in 7 of 38 patients (18.4%), neither of these findings was present in any patient by postsurgery day 42. 

Best-corrected visual acuity scores indicated reduced VA in the study eyes of 7 of 34 patients (20.6%) who received ocriplasmin and no patients in the control group. All of these reductions in VA began on the day of the study injection and most (5/7) resolved, with VA returning to within one Early Treatment Diabetic Retinopathy Study line of the baseline value or better in all 5 patients. The investigator considered these AEs to be mild in intensity and probably related to the study drug. 

Discussion

Characterization of how ocriplasmin functions in the vitreous is important in understanding the process of pharmacologic vitreolysis with ocriplasmin. In this study, the inactivation of ocriplasmin was rapid, as indicated by the high inactivation rate constant of 925 M⁻¹ s⁻¹. Within 2 to 4 hours after a single injection, the mean active ocriplasmin concentration from a 125-μg dose reduced to <25% of the mean active ocriplasmin concentration seen directly after injection (within 5–30 minutes). By 24 hours after injection, the mean active ocriplasmin concentration had reduced to <5% of the mean active ocriplasmin concentration seen directly after injection (within 5–30 minutes). At 7 days after ocriplasmin injection, the mean observed active ocriplasmin concentration from vitreous samples was below the LLQ, comparable to that of the control eyes. 

The concentrations of active ocriplasmin in the vitreous samples varied among patients, particularly in patients with the shortest times from intravitreal injection to vitreous sampling (5–30 and 31–60 minutes). This variation could be explained in part by the small (0.5 mL) volume of vitreous obtained and differences in the sampling location versus the point of injection if a homogeneous solution of ocriplasmin in the vitreous had not yet been obtained. Further, as ocriplasmin degrades rapidly, ^24,25 differences in ocriplasmin activity are not unexpected in samples obtained 5 minutes postinjection compared with samples obtained 30 minutes postinjection, the vitreous sampling window for patients in group 1. However, perhaps of greatest clinical relevance is the fact that active ocriplasmin levels were consistently below the LLQ at 7 days postinjection in all patients. 

These results are consistent with previous studies that demonstrated rapid inactivation of ocriplasmin. In a study of freshly harvested porcine eyes, time-dependent decreases of ocriplasmin (125 μg) activity were observed as well, with minimal active ocriplasmin detected after 24 hours. ^20,26 Furthermore, the inactivation profile of ocriplasmin in pooled human vitreous fluid obtained from random vitrectomy patients and incubated at 37°C showed that less than 0.6% of initial active ocriplasmin was detected at 72 hours. ²⁵  

Formal experiments have not fully characterized the mechanisms leading to ocriplasmin inactivation in the human vitreous, although research has shown that patients with vitreoretinal disorders and those undergoing retinal surgery have fibrinolytic factors present in the vitreous that normally reside in the blood, such as α₂-antiplasmin. ^27,28 Particularly alongside conditions involving blood–retinal barrier breakdown, such as macular pucker or proliferative diabetic retinopathy, α₂-antiplasmin is found in increasing concentrations in the vitreous compared to those in healthy eyes. ^27,29 Alpha₂-antiplasmin is the physiologic inhibitor of plasmin, the full-length protein from which ocriplasmin is derived; the antagonist immediately inactivates plasmin and ocriplasmin in blood. ^22,30 Therefore, α₂-antiplasmin present in the vitreous may contribute to deactivation of injected ocriplasmin. However, the relative contribution of α₂-antiplasmin to the inactivation of ocriplasmin within the vitreous is likely limited. One study evaluating the level of α₂-antiplasmin in the vitreous of patients with unspecified ocular pathology calculated that the amount present (64 nM) would inhibit approximately 4% of the injected dose of ocriplasmin (1500 nM). ²⁴ In contrast, any active ocriplasmin entering the systemic circulation would be rapidly deactivated by α₂-antiplasmin, which is present in high levels within the blood. ³¹ Alternatively or in addition to the previous mechanism of inactivation, in vitro porcine experiments have raised the possibility that ocriplasmin present at therapeutic concentrations in the vitreous could be inactivated through autolytic degradation. ²⁶ Future experiments are required to support these hypotheses. 

The safety results of this study demonstrate that a single intravitreal injection of ocriplasmin (125 μg) is well tolerated at different time points for as long as a week in advance of planned primary vitrectomy. For patients treated with ocriplasmin, the most common AEs were ocular events (90.8%); of these events, 96.6% occurred in the study eye, and almost half occurred postvitrectomy. This is consistent with a single intravitreal injection of a drug that is rapidly inactivated. The most common previtrectomy AEs were reduced VA, vitreous floaters, photopsia, and chromatopsia (dyschromatopsia). The occurrence of such events is thought to be due to the sudden separation of the vitreous from the retina inducing a reversible trauma-like impact on the retina. Most frequently seen in the postoperative follow-up after vitrectomy were increased IOP and macular edema. In the control group, the majority of AEs were ocular events, all of which were in the study eye. The most common ocular AE in the control group was increased IOP. Overall, safety results suggest that no significant hazard is associated with ocriplasmin administration relative to the natural history of the underlying conditions treated and the risk profile of vitrectomy, since most of these events were of mild intensity and resolved. ³²  

As the study investigated only a small number of patients, safety conclusions must be taken within the context of this small size. However, these safety results are generally consistent with data from the two large phase 3 studies of ocriplasmin in patients with symptomatic VMA including when associated with macular hole. ²³ As in the current study, the most common AEs in previous studies were also those known to be associated with vitreous detachment, such as vitreous floaters or photopsia, most of which were transient and mild in severity. Importantly, results from these large trials suggest that there is no increase in serious AEs with ocriplasmin compared with placebo, and no increased risk in terms of acute cataract development or endophthalmitis. ²³  

The results herein, taken together with those from the large ocriplasmin clinical program, support the acceptable safety and tolerability profile of ocriplasmin (125 μg) intravitreal injection. ^23,33  

In conclusion, active ocriplasmin levels in vitreous samples decreased with the time from injection to sample, with active enzyme concentrations decreasing to below the LLQ (<272.4 ng/mL) after at least 24 hours. In the 7-day cohort, all vitreous samples revealed ocriplasmin levels to be below the LLQ, similar to controls. The study also further supports the acceptable safety and well-tolerated profile of ocriplasmin (125 μg) observed throughout the phase 2/3 clinical development program. 

Acknowledgments

Data analysis and interpretation were provided by Bart Jonckx of ThromboGenics NV. Technical and editorial assistance was provided by Diane Kwiatkoski, Karen Munro, and Rebecca A. Bachmann of Quintiles Communications, Parsippany, New Jersey. 

Supported by ThromboGenics NV, including a research grant (PS). 

Disclosure: P. Stalmans, ThromboGenics NV (F, R), Ellex (R), Dutch Ophthalmic Research Center (R), Bausch & Lomb (F); A. Girach, ThromboGenics NV (I, E)