ONS-5010 (bevacizumab-vikg) Safety and Efficacy in Subfoveal Choroidal Neovascularization Secondary to Age-related Macular Degeneration
Abstract
Background and Objective
This was a prospective multicenter, randomized, double-masked, active-controlled study, the aim of which was to demonstrate the efficacy and safety of intravitreal ONS-5010 (bevacizumab-vikg) in eyes with neovascular age-related macular degeneration (nAMD). This was a phase III trial on ONS-5010 (NORSE TWO).
Materials and Methods
Treatment-naïve nAMD patients aged 50 years and older with a best-corrected distance visual acuity (BCVA) of 25 to 67 Early Treatment Diabetic Retinopathy Study (ETDRS) letters and evidence of disease activity were included. Subjects randomized to ONS-5010 received monthly intravitreal doses of 1.25 mg of ONS-5010, bevacizumab-vikg (Outlook Therapeutics) for 12 months. Subjects randomized to ranibizumab received 0.50 mg of ranibizumab on days 0, 30, 60, 150, and 240 based on the PIER study dosing regimen.
Results
The primary end point was the proportion of subjects who gained ≥ 15 letters from baseline in BCVA at 11 months, and evaluating the safety and tolerability of intravitreal injections of ONS-5010 administered monthly from baseline to 12 months. One hundred thirteen subjects were included in the ONS-5010 group and 115 subjects were included in the ranibizumab group. Respectively, 41.7% and 23.1% of patients gained ≥ 15 letters (3 lines) of visual acuity, with a risk difference of 0.1859 [95% CI = 0.0442, 0.3086]; P = 0.0052. The change in BCVA from baseline to 11 months was found to be 11.2 ± 12.19 and 5.8 ± 14.80 ETDRS letters, respectively. The number of subjects gaining ≥ 5 and ≥ 10 letters and subjects losing < 15 letters was significantly higher in the ONS-5010 group. Similarly, subjects with a Snellen visual acuity equivalent of 20/200 (35 ETDRS letters) or worse at 11 months were significantly fewer in the ONS-5010 group. Only one subject in the ONS-5010 group had a study-related serious ocular adverse event (SAE), namely, elevated intraocular pressure. The most common adverse event in the ONS-5010 group was conjunctival hemorrhage (8.8%), and reduced visual acuity in the ranibizumab group (12.2%).
Conclusions
In the prescribed treatment plan, ONS-5010 exhibited strong effectiveness in improving or stabilizing visual acuity and was also well tolerated. Bevacizumab and ranibizumab displayed a comparable safety profile.
[Ophthalmic Surg Lasers Imaging Retina 2024;55:XX–XX.]
Introduction
Globally, one of the largest causes of blindness in the elderly population is neovascular age-related macular degeneration (nAMD). According to the World Health Organization, over 200 million people suffer from nAMD1 and that number is expected to rise as life span increases.2
Neovascular AMD causes a rapid loss of central vision as a result of macular blood and/or serous fluid exudation, which itself arises from choroidal neovascularization (CNV) under the macula.3 In turn, CNV is formed due to increased levels of vascular endothelial growth factor-A, a cytokine that is secreted in response to hypoxia.4 This cytokine also increases exudation from blood vessels,5,6 which can progress to submacular fibrosis and result in permanent loss of central vision.2
In 2004, the United States Food and Drug Administration (US FDA) approved pegaptanib as the first intravitreal anti-vascular endothelial growth factor (anti-VEGF).6 The treatment of nAMD was then revolutionized by the subsequent introduction of newer and more effective anti-VEGFs, such as ranibizumab, bevacizumab, and aflibercept.6 In 2006, ranibizumab was FDA-approved as Lucentis for the treatment of nAMD. However, despite the impact that anti-VEGFs have on visual outcomes in nAMD, the necessity of repeated injections and the high cost of treatment remain a barrier to compliance.6
Bevacizumab is a humanized monoclonal antibody that inhibits VEGF 7 and was FDA-approved as Avastin in 2004 for the treatment of colorectal cancer.8 It is derived from the same monoclonal antibody as ranibizumab,6 and was initially used off-label in cases of nAMD while ranibizumab was under consideration for FDA approval but, since then, randomized trials have shown that bevacizumab in fact has the same target specificity9–11 as ranibizumab, with similar safety profile and efficacy and a lower cost.12,13 Confirming these findings was the aim of the US CATT and the United Kingdom (UK) IVAN trials14,15 and, through the latter, it was found that both Avastin and Lucentis are of comparable safety and efficacy in nAMD when administered monthly or on an as-needed basis (PRN).15 The IVAN trial showed that bevacizumab and ranibizumab have similar safety and efficacy, but bevacizumab (as Avastin) may be more cost effective.12,16
Due to its safety and efficacy and relatively lower cost, Avastin is still used off-label in the treatment of neovascular eye diseases, after aliquoting a single-use vial, which can be potentially unsafe.7,17 In order to be used for intravitreal injections, multiple single-dose packages are obtained from each Avastin vial. This packaging is not approved for ophthalmic use and storage in this packaging may cause protein aggregation and reduce drug efficacy, besides increasing the risk of infectious and noninfectious endophthalmitis,17,18 which can occur as a result of improper storage or handling.19 Differences in the composition of repackaged bevacizumab, as well as between repackaged and reference bevacizumab, have been documented; specifically, there are notable differences in particulate density.20 The accumulation of particulate matter in repackaged bevacizumab over time could potentially impact the safety of off-label Avastin. Moreover, underdosing is common with repackaging, and this can give a false impression that the patient is not responding to the treatment.17 Hence, an ophthalmic preparation of bevacizumab is preferred, given that this has the potential to improve safety and efficacy.
ONS-5010 is an ophthalmic-specific preparation of bevacizumab in development for the treatment of retinal diseases such as nAMD, diabetic macular edema (DME), and branch retinal vein occlusion (BRVO). The current study is the pivotal phase III trial (NORSE TWO) designed to assess the efficacy and safety of ONS-5010 for the treatment of nAMD.
Methods
Study Design
The study is a prospective, multicenter, randomized, double-masked, controlled study designed to evaluate the efficacy and safety of intravitreally administered ONS-5010. This was a phase III trial of ONS-5010 (NORSE TWO).
Two hundred twenty-eight eligible subjects with neovascular (wet) age-related macular degeneration (nAMD) were enrolled at 39 study centers in the United States. The study adhered to the tenets of the Declaration of Helsinki and was approved by an institutional review board. Each participant gave written informed consent.
Following a screening period of up to 28 days, eligible subjects were randomized in a 1:1 ratio to receive either ONS-5010 or ranibizumab in the study eye. Only one eye per patient was designated as the study eye. The injection was performed by an unmasked physician. Prior to randomization, the investigators reviewed clinical laboratory tests for eligibility and received confirmation of subject eligibility from the medical image reviewer.
Patient Selection
Eligible subjects for the study included patients with nAMD, aged at least 50 years, with previously untreated active subfoveal CNV lesions secondary to AMD, with or without a classic CNV component, shown via active leakage on fluorescein angiography and/or central retinal edema on spectral-domain optical coherence tomography (SD-OCT). Best-corrected distance visual acuity (BCVA) was required to be 25 to 67 Early Treatment Diabetic Retinopathy Study (ETDRS) letters. If both eyes were eligible, the eye with more advanced disease was selected, based on the investigator/sponsor's discretion. The fellow eye should have a BCVA of at least 20 ETDRS letters and may or may not have been previously treated with anti-VEGFs.
Exclusion criteria included: previous use of intravitreal corticosteroids or anti-VEGFs (including Avastin) in the study eye; prior treatment of the study eye with verteporfin, external beam radiation therapy, transpupillary thermotherapy, subfoveal focal laser photocoagulation, vitrectomy or subfoveal surgery; concurrent ocular disease in the study eye that would require treatment during the study duration; history of rhegmatogenous retinal detachment in the study eye; history of idiopathic or autoimmune-associated uveitis, infectious conjunctivitis, keratitis, scleritis or endophthalmitis in either eye; treatment of the fellow eye with verteporfin within 7 days prior to randomization; and any previous participation in any investigational drug study (other than vitamins and minerals) within one month prior to randomization.
Data Collection
Safety and efficacy assessments were conducted throughout the dosing and evaluation phases of the study. The efficacy evaluation period was 11 months (through day 330) for both study groups, whereas the safety evaluation period was 11 months (through day 330) for the ranibizumab group and 12 months (through day 365) for the ONS-5010 group.
Efficacy was determined by BCVA, while safety was assessed by reviewing adverse events (AEs): postinjection assessments (including gross visual assessments [finger counting, hand motion, and light perception] 15 minutes postinjection and intraocular pressure [IOP] measurements 30 minutes postinjection), review of body systems, vital sign measurements, clinical laboratory test results, and follow-up measurements of IOP, slit-lamp biomicroscopy, dilated ophthalmoscopy, fundus autofluorescence, and fluorescein angiography. All investigators performing the ocular assessments were masked to the study treatment assignment.
Study Treatment
The treatment under investigation, ONS-5010, contains recombinant, humanized, anti-VEGF monoclonal antibody (bevacizumab-vikg), and is being developed for intravitreal injection in patients with nAMD.
The control treatment, ranibizumab (Lucentis), is approved for the treatment of nAMD, macular edema following retinal vein occlusion, diabetic macular edema, diabetic retinopathy, and myopic CNV.
Treatment Protocol
Subjects randomized to receive ONS-5010 received a monthly intravitreal injection of 1.25 mg of ONS-5010 in the study eye for up to 12 months until study completion. Subjects randomized to ranibizumab received 0.50 mg of ranibizumab via intravitreal injection in the study eye every month for 3 months (ie, on days 0, 30, and 60) followed by two additional injections, 90 days apart, on days 150 and 240, based on the PIER study dosing regimen.21 Subjects in the ranibizumab group underwent sham procedures during visits when they did not receive an active (ranibizumab) injection (Figure 1). The sham procedures were intended to maintain patient masking. The patient was prepped as if they would receive an actual intravitreal injection and the eye was touched by a syringe without a needle.
Study scheme.
The last visit of the study differed depending on which treatment group the subject was randomized to, concluding at day 330 for the ranibizumab group and at day 365 for the ONS-5010 group. At this time point, the subjects returned to the investigators' standard of care. Patients who experienced a loss of ≥ 15 letters (3 lines) compared to their baseline at any visit, were provided with a rescue therapy as determined by the investigator.
Outcome Measures
The primary objectives of the study were to evaluate the efficacy of intravitreal injections of ONS-5010 with respect to ranibizumab in preventing vision loss (as measured by the difference in proportion of subjects who gain ≥ 15 letters from baseline in BCVA at 11 months) as well as the safety and tolerability of intravitreal injections of ONS-5010 administered monthly from baseline to 12 months.
The secondary objective for the study was to compare the efficacy of intravitreal injections of ONS-5010 with respect to ranibizumab in preventing vision loss or gaining visual acuity, as measured by: the mean change in BCVA from baseline to 11 months; responder analysis including the proportion of subjects who gain ≥ 5 or ≥ 10 letters in visual acuity at 11 months compared with baseline; the proportion of subjects who lose fewer than 15 letters in visual acuity at 11 months compared with baseline; and the proportion of subjects with Snellen visual acuity equivalent to 20/200 or worse at 11 months.
Safety was evaluated by monitoring ocular and nonocular AEs, serious adverse events (SAEs), clinical laboratory findings, vital sign measurements, ophthalmic examinations, and a review of concomitant medications.
Statistical Methods
All statistical processing was performed using SAS version 9.4 (SAS Analytics Software) or later. Descriptive statistics were used to provide an overview of the efficacy and safety results. Means, standard deviations (SD), standard errors of the mean, and 95% confidence intervals (CIs) were calculated where possible. Two-sided 95% CIs were provided for all endpoints when applicable. A two-sided P value of < 0.05 was considered statistically significant. The primary and secondary efficacy analyses were calculated using a Fisher's exact test. The first secondary endpoint (ie, the mean change in ETDRS letters from baseline to 11 months) was calculated using least square (LS) mean fixed and adaptive approaches. Additionally, complementary analyses regarding the first secondary endpoint included an analysis of covariance (ANCOVA) model, with the study treatment group as a fixed effect and baseline BCVA as a continuous covariate, an ANCOVA employing a multiple imputation method, and a repeated measures analysis of the change in BCVA from baseline over time.
Analysis of the results was based on the intention-to-treat population (ITT), but statistics for the per-protocol population (PP) were also calculated to confirm findings. All AEs occurring during the study were recorded and classified using the Medical Dictionary for Regulatory Activities (version 23.0) and AE severity was graded using the National Cancer Institute Common Terminology Criteria for AEs (CTCAE) (version 4.03).
Participants who stopped the study because of adverse effects, ineffectiveness, or the need for rescue treatment were considered as nonresponders. Those who left the study for other reasons and missed the BCVA measurement at month 11 were considered missing.
Results
Patient Demographics
A total of 228 subjects were randomized at 39 study centers across the US. Of those, 198 (86.8%) completed the study. The three most frequently reported reasons for study discontinuation were withdrawal of consent (4.8%), AEs (3.5%), and other reasons (1.8%).
All 228 randomized subjects were included in the ITT and safety populations (113 subjects in the ONS-5010 group and 115 subjects in the ranibizumab group), and 163 subjects were included in the PP population (85 and 78 subjects in the ONS-5010 and ranibizumab groups, respectively). In the ONS-5010 group, nine subjects discontinued the treatment and withdrew from the study (five because of AEs, three lost to follow-up, and one withdrew voluntarily); one subject completed treatment but withdrew from the study; and two subjects discontinued treatment but completed the study. In the ranibizumab group, 20 subjects discontinued the treatment and withdrew from the study (10 due to voluntary withdrawal, five because of AEs, three because of lack of effectiveness, and three for other reasons); one subject discontinued treatment but completed the study. Treatment-emergent adverse events (TEAEs) occurred in three cases in the ONS-5010 group and four cases in the ranibizumab group as a result of COVID-19. On the whole, the pandemic resulted in a considerable number of withdrawals from the study either due to TEAEs or due to fear of infection.
Table 1 presents the demographic characteristics of the ITT/safety population. Nine subjects with prior anti-VEGF treatment in the study eye were included in the trial before the protocol was amended to exclude previous treatment of the study eye with anti-VEGF.
| Parameter | Ranibizumab (n = 115) | ONS-5010 (n = 113) | Overall (n = 228) |
|---|---|---|---|
| Sex, n (%) | |||
| Male | 46 (40.0%) | 46 (40.7%) | 92 (40.4%) |
| Female | 69 (60.0%) | 67 (59.3%) | 136 (59.6%) |
| Age, mean ± SD | 79.1 ± 8.49 | 78.8 ± 8.30 | 78.9 ± 8.38 |
| Race, n (%) | |||
| White | 113 (98.3%) | 110 (97.3%) | 223 (97.8%) |
| Asian | 1 (0.9%) | 1 (0.9%) | 2 (0.9%) |
| Black | 0 (0%) | 0 (0%) | 0 (0%) |
| Unreported | 1 (0.9%) | 0 (0%) | 1 (0.4%) |
| Study eye, n (%) | |||
| Right | 51 (44.3%) | 49 (43.4%) | 100 (43.9%) |
| Left | 64 (55.7%) | 64 (56.6%) | 128 (56.1%) |
| Prior anti-VEGF, n (%) | 5 (4.3%) | 4 (3.5%) | 9 (3.9%) |
| Baseline BCVA, mean ± SD | 51.1 ± 12.96 | 52.1 ± 12.16 | 51.6 ± 12.55 |
| ≤ 54 letters, n (%) | 55 (47.8%) | 49 (43.4%) | 104 (45.6%) |
| ≥ 55 letters, n (%) | 60 (52.2%) | 64 (56.6%) | 124 (54.4%) |
| Baseline CFT (μm), mean ± SD | 423.7 ± 114.77 | 430.0 ± 150.85 | 426.8 ± 133.61 |
Primary Efficacy Outcomes
The proportion of subjects achieving an increase of ≥ 15 ETDRS letters in BCVA from baseline to 11 months for the ONS-5010 group was 41.7% (n = 45/108) and 23.1% (n = 24/104) for the ranibizumab group, with a risk difference of 0.1859 (95% CI = 0.0442, 0.3086); P = 0.0052. Figure 2 shows the percentage of subjects who gained ≥ 15 ETDRS letters in BCVA over time.
Primary efficacy endpoint: percentage of subjects gaining ≥15 Early Treatment Diabetic Retinopathy Study (ETDRS) letters from baseline in best-corrected visual acuity (BCVA) over time (intent-to-treat [ITT] population). CI = confidence interval.
Secondary Efficacy Outcomes
The change in BCVA from baseline to 11 months was found to be 11.2 ± 12.19 ETDRS letters (range: −47 to 40) in the ONS-5010 group and 5.8 ± 14.80 ETDRS letters (range: −42 to 42) in the ranibizumab group. When calculated using the trimmed means fixed approach, the least squares (LS) mean change in BCVA from baseline to 11 months was 19.310 letters in the ONS-5010 group and 14.575 letters in the ranibizumab group (LS mean difference of 4.735 [95% CI = 1.306, 8.163]; P = 0.0035). When calculated using the trimmed means adaptive approach, the LS mean change in BCVA from baseline to 11 months was 13.640 letters in the ONS-5010 group and 5.770 letters in the ranibizumab group. The difference observed between study treatment groups was significant, with an LS mean difference of 7.870 (95% CI = 4.536, 12.599; P = 0.0001).
The first and second complementary analyses of the secondary efficacy endpoint, which used ANCOVA and an ANCOVA with multiple imputation, consistently showed a greater improvement in BCVA in the ONS-5010 group compared with the ranibizumab group at 11 months (P = 0.0043 for ANCOVA and P = 0.0014 for ANCOVA with multiple imputation) (Table 2). The third complementary analysis, which used repeated measures analysis of the first secondary efficacy endpoint, showed a similar improvement in BCVA in both groups as early as day 30, with a greater improvement in the ONS-5010 group than the ranibizumab group, starting at day 150 and being maintained through day 300 in the ITT population (P ≤ 0.0272 at each time point). As shown in Figure 3, primary response (the first 3 months) was similar in both groups. However, the improvement observed in the secondary response was maintained in the ONS-5010 group, whereas the mean gain of ETDRS letters reduced in the ranibizumab group after 120 days.
| CDVA Change From Baseline to 11 Months | Ranibizumab (n= 115), n= 96 | ONS-5010 (n= 113), n= 104 |
|---|---|---|
| Mean (SD) | 5.8 (14.80) | 11.2 (12.19) |
| Median | 6.0 | 12.0 |
| Min, max | −42, 42 | −47, 40 |
| ANCOVA | ||
| LS mean (SE) | 5.807 (1.3519) | 11.226 (1.2989) |
| 95% CI | 3.141, 8.473 | 8.665, 13.788 |
| LS mean difference | 5.419 (1.8748) | |
| 95% CI | 1.722, 9.117 | |
| P value | 0.0043 | |
| ANCOVA with multiple imputation | ||
| LS mean (SE) | 5.131 (1.3632) | 11.191 (1.3112) |
| 95% CI | 2.459, 7.803 | 8.621, 13.761 |
| LS mean difference | 6.060 (1.8927) | |
| 95% CI | 2.350, 9.770 | |
| P value | 0.0014 |
Third complementary analysis of mean change in best-corrected visual acuity (BCVA) in Early Treatment Diabetic Retinopathy Study (ETDRS) letters from baseline over time by repeated measures analysis (intent-to-treat [ITT] population). SE = standard error.
The remaining secondary end points are summarized in Table 3. Seventy-four subjects (68.5%) gained ≥ 5 ETDRS letters from baseline to 11 months in the ONS-5010 group vs 53 (51.0%) in the ranibizumab group (95% CI = 0.0315, 0.3052; P = 0.01). Similarly, 61 subjects (56.5%) gained ≥ 10 ETDRS in the ONS-5010 group vs 36 (34.6%) in the ranibizumab group (95% CI = 0.0726, 0.3487; P = 0.0016). The number of subjects losing < 15 letters from baseline at 11 months was 101 (93.5%) in the ONS-5010 group and 86 (82.7%) in the ranibizumab group (95% CI = 0.0168, 0.2044; P = 0.018). Only 14 subjects (13%) in the ONS-5010 group and 25 (24%) subjects in the ranibizumab group had a visual acuity equivalent of 20/200 Snellen or worse at 11 months (95% CI = −0.2187, −0.0050; P = 0.05). More than 80% of patients in the ONS-5010 group maintained BCVA each month and at one year, and 86.4% maintained or gained letters.
| Parameter, n/N (%) | Ranibizumab (n = 115) | ONS-5010 (n = 113) |
|---|---|---|
| Subjects gaining ≥5 letters from baseline at 11 months | 53/104 (51.0) | 74/108 (68.5) |
| Risk difference | 0.1756 | |
| 95% CI | 0.0315, 0.3052 | |
| P value | 0.0116 | |
| Subjects gaining ≥10 letters from baseline at 11 months | 36/104 (34.6) | 61/108 (56.5) |
| Risk difference | 0.2187 | |
| 95% CI | 0.0726, 0.3487 | |
| P value | 0.0016 | |
| Subjects losing <15 letters from baseline at 11 months | 86/104 (82.7) | 101/108 (93.5) |
| Risk difference | 0.1083 | |
| 95% CI | 0.0168, 0.2044 | |
| P value | 0.0185 | |
| Subjects with a visual acuity Snellen equivalent of 20/200 or worse at 11 monthsa | 25/104 (24.0) | 14/108 (13.0) |
| Risk difference | −0.1108 | |
| 95% CI | −0.2187, −0.0050 | |
| P value | 0.0505 |
Safety Outcomes
The majority of TEAEs observed in the subjects were not considered by the investigators to be study treatment/study procedure–related (Table 4). The frequency of TEAEs considered by the investigators to be related to the study treatment/study procedure was greater in the ONS-5010 group compared with the ranibizumab group (18.6% [21 subjects] vs 7.8% [9 subjects]). Hypothetically, this may be due to a greater number of injections in the ONS-5010 group (12 injections in the ONS-5010 group versus five injections in the ranibizumab group).
| Criteria | Ranibizumab (n = 115), n (%) | ONS-5010 (n = 113), n (%) |
|---|---|---|
| At least 1 TEAE | 85 (73.9%) | 85 (75.2%) |
| At least 1 TEAE related to study drug/study procedure | 9 (7.8%) | 21 (18.6%) |
| At least 1 ocular TEAE | 61 (53.0%) | 59 (52.2%) |
| At least 1 ocular TEAE related to study drug/study procedure | 8 (7.0%) | 21 (18.6%) |
| At least 1 ≥ grade 3 TEAE | 13 (11.3%) | 14 (12.4%) |
| At least 1 ≥ grade 3 TEAE related to study drug/study procedure | 1 (0.9%) | 2 (1.8%) |
| At least 1 serious TEAE | 16 (13.9%) | 14 (12.4%) |
| At least 1 serious TEAE related to study drug/study procedure | 1 (0.9%) | 2 (1.8%) |
| At least 1 TEAE leading to discontinuation | 5 (4.3%) | 2 (1.8%) |
Over the course of the study, 30 subjects (13.2%) experienced one or more treatment-emergent SAEs each (fatal and nonfatal combined). This included 14 subjects (12.4%) in the ONS-5010 group and 16 subjects (13.9%) in the ranibizumab group. Only one subject in the ONS-5010 had a study-related ocular SAE, namely, elevated IOP.
Within the ONS-5010 group, the ocular events occurring in the study eye reported at the three highest frequencies were conjunctival hemorrhage (8.8% [10 subjects]) and elevated IOP (6.2% [seven subjects]), followed by nuclear cataract, corneal abrasion, retinal hemorrhage, reduced visual acuity, vitreous detachment, and vitreous floaters (3.5% [four subjects]). Notably, only one case of ocular inflammation (iritis) was reported in the study eyes of the ONS-5010 subjects and no case of vasculitis was recorded.
Within the ranibizumab group, the ocular events occurring in the study eye reported at the three highest frequencies were reduced visual acuity (12.2% [14 subjects]), retinal hemorrhage (5.2% [six subjects]), and dry eye (4.3% [five subjects]). No treatment-related ocular SAE was recorded.
A total of five subjects died, including three in the ONS-5010 group and two in the ranibizumab group. The events that resulted in death were COVID-19 and respiratory failure (both occurring in the same subject of the ONS-5010 group), myocardial infarction and death due to COVID-19 (each occurring in different subjects of the ONS-5010 group), and cardiac arrest and cardiac death (each occurring in different subjects in the ranibizumab group). These events were not considered by the investigators to be related to the study treatment/study procedure.
Discussion
This randomized double-masked controlled study was a pivotal phase III study (NORSE TWO) and was designed to evaluate the efficacy and safety of ONS-5010 dosed monthly versus ranibizumab dosed for three initial monthly injections, followed by quarterly dosing based upon the PIER study dosing regimen, in adult subjects with active primary subfoveal CNV, with or without a classic CNV component, secondary to AMD. While clinicians often treat nAMD with greater frequency than the ranibizumab arm of the NORSE TWO study, the study was designed to meet a regulatory endpoint of comparison to a previously approved treatment dosing regimen in order to demonstrate the existence of a treatment effect for ONS-5010.
Unlike the previous studies, such as CATT, that used commercially available 22 repackaged bevacizumab, this is the first study to use an ophthalmic preparation of bevacizumab. The study recruited 228 subjects who were randomized at 39 US study centers. Of those 228 subjects, 198 (86.8%) completed the study. NORSE TWO achieved the goal of demonstrating the existence of an effect of ONS-5010 by achieving statistical significance in the primary endpoint of the proportion of subjects achieving an increase of ≥ 15 letters in BCVA from baseline to 11 months, with 41.7% and 23.1% being recorded in the ONS-5010 and ranibizumab groups, respectively, with a risk difference of 0.1859 (95% CI = 0.0442, 0.3086; P = 0.0052).
Furthermore, a significant distinction was observed between the study treatment groups regarding secondary efficacy endpoints. ONS-5010 showed a favorable outcome compared to ranibizumab for the first secondary efficacy endpoint, which measured the change in BCVA from baseline to 11 months. This outcome was determined using two different trimmed means approaches. Specifically, when employing the trimmed means fixed approach, the LS mean change in BCVA from baseline to 11 months was 19.310 letters in the ONS-5010 group and 14.575 letters in the ranibizumab group. This resulted in an LS mean difference of 4.735 (95% CI = 1.306, 8.163; P = 0.0035). Likewise, when utilizing the trimmed means adaptive approach, the LS mean change in BCVA from baseline to 11 months was 13.640 letters in the ONS-5010 group compared to 5.770 letters in the ranibizumab group. This yielded an LS mean difference of 7.870 (95% CI = 4.536, 12.599; P = 0.0001).
Moreover, ONS-5010 showed excellent efficacy in terms of the secondary endpoints (Table 2) (ie, the proportions of subjects who gained ≥ 5 letters and ≥ 10 letters in BCVA) and also enabled a high number of subjects to maintain their BCVA. In the ONS-5010 group, 74 subjects (68.5%) gained ≥ 5 letters in BCVA, 61 subjects (56.5%) gained ≥ 10 letters, and 101 subjects (93.5%) lost < 15 letters from baseline to 11 months. In the ranibizumab group, 53 subjects (51.0%), 36 subjects (34.6%), and 86 subjects (82.7%) were in the respective categories. For each of these fixed-sequence secondary endpoints, the difference between study treatment groups was significant (P values range: 0.05 to 0.0016).
The proportion of subjects with a Snellen visual acuity equivalent of 20/200 or worse at 11 months (ie, the final secondary efficacy endpoint in the fixed sequence) was lower in the ONS-5010 group (13.0% [14 subjects]) with respect to the ranibizumab group (24.0% [25 subjects]). However, this difference was not statistically significant (P = 0.0505).
The complementary analyses of the secondary efficacy endpoints using ANCOVA and ANCOVA with multiple imputation consistently showed a greater improvement in BCVA in the ONS-5010 group compared with the ranibizumab group at 11 months (P ≤ 0.0043). The repeated measures analysis of the first secondary efficacy endpoint showed an improvement in BCVA in both groups as early as day 30, with a greater improvement being observed in the ONS-5010 group with respect to the ranibizumab group starting at day 150 and maintaining through day 300 in the ITT population (P ≤ 0.0272 at each time point). Additionally, ≥ 80% of eyes maintained BCVA each month while 86.4% maintained BCVA or gained ETDRS letters at 12 months, indicating the sustained effect of ONS-5010.
The efficacy analyses conducted using the PP population were generally similar to the analyses conducted using the ITT population, further confirming the efficacy conclusions of the study.
Regarding safety, the frequency of TEAEs that were considered by the investigators to be related to the procedure or the study treatment was higher in the ONS-5010 group compared with the ranibizumab group (18.6% vs 7.8%, respectively). However, the number of SAEs that were considered by the investigators to be related to the procedure or the study treatment was comparable in both groups (two subjects vs one subject, respectively). It should be noted that the number of injections in the ONS-5010 group was more than double the number of injections in the ranibizumab group.
Previous studies that compared bevacizumab to ranibizumab, such as the CATT study, the IVAN study, and the BRAMD study, in addition to a Cochrane systematic review, proved that bevacizumab is noninferior to ranibizumab when both are used in a similar dose regimen.16,22–25 Additionally, recent meta-analysis has shown that bevacizumab displays similar efficacy to other anti-VEGFs such as ranibizumab and brolucizumab.26 These findings are consistent with the results of this study in the first 4 months when both drugs were administered in a similar dosing regimen.
By comparing the bevacizumab monthly group in the CATT study to the bevacizumab group in the current study, 94% of subjects in the bevacizumab monthly group in the CATT study lost < 15 letters, compared to 93.5% in the current study. In the CATT study, 31.3% gained ≥ 15 letters with a mean gain of eight letters.22 In the current study, 41.7% gained ≥ 15 letters with a mean gain of 11.2 letters. It should be noted that the bevacizumab used in the CATT study was commercially acquired and repackaged in glass vials under aseptic conditions.22 One reason for this discrepancy could be the lower potency of repackaged bevacizumab, which may affect its overall effectiveness.
When the initial (first 3 months) and secondary responses were compared (Figure 2), both treatments displayed a similar initial response when a similar dosing regimen (monthly injections) was used, whereas the bevacizumab group gained more letters in the subsequent months compared to the ranibizumab group. The dosing regimen in the ranibizumab group is the likely reason for differentiation in study arms at later timepoints.6,27,28
Since the first 3 months of the current study are consistent with the monthly groups of the CATT study, it is interesting to review the initial response of the monthly bevacizumab and ranibizumab groups. At week 12 of the CATT study, 21% in the monthly bevacizumab group gained ≥ 15 ETDRS letters vs 18% in the ranibizumab monthly group, 75% vs 79% had a change within ±14 letters, and 5% vs 3% lost ≥ 15 letters. In the CATT study, both groups gained 6.1 ± 0.7 letters at 12 weeks whereas, in the current study, the ranibizumab group gained 8 ± 11.7 letters and the bevacizumab group gained 9.2 ± 10.7 letters (P = 0.4).22 While bevacizumab and ranibizumab have shown comparable behavior both in the CATT study and the current study, the greater mean gain of ETDRS letters in the current study is probably due to the worse baseline corrected distance visual acuity in this study (51.1 ± 12.96 and 52.1 ± 12.16 ETDRS for ranibizumab and bevacizumab, respectively) than in the CATT study (60.1 ± 14.3 and 60.2 ± 13.1 ETDRS letters, respectively).29
Over the course of the study, 30 subjects (13.2%) experienced one or more treatment-emergent SAEs each (fatal and nonfatal combined): 14 subjects (12.4%) in the ONS-5010 group and 16 subjects (13.9%) in the ranibizumab group. Only one study-related ocular SAE was detected in the ONS-5010 group, namely, elevated IOP, and no study-related ocular SAE was reported in the ranibizumab group.
Similarly, there were no important differences between the ONS-5010 and ranibizumab groups regarding the percentages of subjects with at least one ocular TEAE (52.2% vs 53.0%, respectively). Notably, across all of the three clinical trials where ONS-5010 was administered, including 486 subjects, only one case (< 1%) of ocular inflammation AE was reported (iritis).30
Anti-VEGFs have drastically improved the visual outcomes in nAMD patients. However, their high cost constitutes an economic burden due to the need for repeated injections,6 considering that nAMD is one of the most common causes of blindness in the elderly population.2 Off-label bevacizumab is considered to be the most cost-effective anti-VEGF,2 but its use through aliquoting single-use vials has been subjected to scrutiny in terms of quality, consistency, and safety. In 2015, Yannuzzi et al. reported significant changes in protein concentrations that appear to be lower than bevacizumab obtained directly from Genentech.17 Additionally, the long-term storage of the solution in certain types of syringes made from materials with an affinity for the protein present in bevacizumab can reduce its concentration in the repackaged dose. For example, plastic syringes can lead to protein aggregations and contamination with silicone oil microdroplets.17,31 Similarly, protein aggregation can occur with glass syringes.32 Such factors can lead to a general decrease in the amount of protein present in compounded samples, as well as significant differences between specimens.17
High molecular weight aggregates introduced during the repackaging process along with the potential shearing/fragmentation of the protein caused by that process can lead to significant differences in immunoglobulin G (IgG) concentration.32 Large particulate matter within syringes may lead to obstruction of aqueous outflow and an elevation in IOP.32 Moreover, functional degradation may occur due to long-term storage of the medication in immediate-use syringes and in suboptimal conditions, affecting the quality of the protein and reducing its concentration. The tendency of aliquoted bevacizumab's activity and efficacy to decrease over time, even if it has been stored at a temperature of 4° C, has been demonstrated.17,33
There have also been reports of infectious and inflammatory endophthalmitis outbreaks related to the intravitreal use of Avastin in the US, Japan, and South Korea, as well as other countries.17,18,34–36 Infectious endophthalmitis can be attributed to contamination and infection resulting from poor aseptic technique or mishandling of the syringes.37,38 It is difficult to attribute a root cause to sterile or noninfectious endophthalmitis, but causative issues may include protein aggregation, or change in conformation, or undetected endotoxins.39 These unfavorable characteristics could be reduced or eliminated if bevacizumab is adapted for ophthalmic use, and ONS-5010 is the first ophthalmic preparation of bevacizumab that has the potential to be FDA-approved.
Using an ITT population is one of the strengths of the study given that it provides more realistic data with respect to a PP population. Nevertheless, PP population statistics confirmed the findings observed using the ITT population.
The different dosing regimen between the two groups is not considered a limitation of the study given that the aim of the study was to demonstrate the effect of bevacizumab. One reason for spacing doses in the ranibizumab group was to replicate a more realistic clinical dosing schedule. Furthermore, according to both the IVAN and CATT study, spaced dosing of ranibizumab was shown to be as effective as monthly dosing.16,22 Notably, this should not compromise the integrity of the results generated in this study. According to the FDA, demonstrating superiority over a concurrent control group, be it an active agent (such as a lower dose of the test treatment) or a placebo, typically offers compelling proof of effectiveness namely because a superiority design does not rely on assumptions regarding the control's effectiveness.40
The study outcome provides strong evidence of the clinical effect of ONS-5010. The ONS-5010 group demonstrated a statistically significant gain in ETDRS letters in multiple assessments of BCVA.
Conclusion
Based on a review of TEAEs, SAEs, and the study treatment/study procedure–related TEAEs, the safety profile of ONS-5010 (administered monthly for up to 12 months) was similar to that of ranibizumab (dosed less frequently) in adult subjects aged 50 years or older with subfoveal CNV secondary to AMD. Overall, the safety findings related to ONS-5010 were expected and the study treatment was safe and well tolerated in this study.
- 1.. World report on vision. October 8, 2019. Accessed June 26, 2023. https://www.who.int/publications/i/item/9789241516570 > Google Scholar
- 2.. Neovascular age-related macular degeneration (nAMD): a review of emerging treatment options. Clin Ophthalmol. 2022; 16:917–933.
10.2147/OPTH.S231913 PMID:35368240 > Crossref MedlineGoogle Scholar - 3.. Age-related macular degeneration histopathologic studies. The 1992 Lorenz E. Zimmerman Lecture. Ophthalmology. 1993; 100(10):1519–1535.
10.1016/S0161-6420(93)31466-1 PMID:7692366 > Crossref MedlineGoogle Scholar - 4.. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science (1979). 1989; 246(4935):1306–1309.
10.1126/science.2479986 > CrossrefGoogle Scholar - 5.. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science (1979). 1983; 219(4587):983–985.
10.1126/science.6823562 > CrossrefGoogle Scholar - 6.. Vascular endothelial growth factor antagonists: promising players in the treatment of neovascular age-related macular degeneration. Drug Des Devel Ther. 2021; 15:2653–2665.
10.2147/DDDT.S295223 PMID:34188445 > Crossref MedlineGoogle Scholar - 7.. Intravitreal administration of bevacizumab: pros and cons. Daru. 2015; 23(1):27.
10.1186/s40199-015-0110-0 PMID:25896903 > Crossref MedlineGoogle Scholar - 8.. Avastin (bevacizumab) information. June 29, 2011. Accessed June 19, 2023. https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/avastin-bevacizumab-information > Google Scholar
- 9.. Optical coherence tomography findings after an intravitreal injection of bevacizumab (Avastin) for neovascular age-related macular degeneration. Ophthalmic Surg Lasers Imaging. 2005; 36(4):331–335.
10.3928/1542-8877-20050701-14 PMID:16156152 > LinkGoogle Scholar - 10.. Intravitreal bevacizumab (Avastin) for neovascular age-related macular degeneration. Ophthalmology. 2006; 113(3):363–372.e5.
10.1016/j.ophtha.2005.11.019 PMID:16458968 > Crossref MedlineGoogle Scholar - 11.. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011; 364(20):1897–1908.
10.1056/NEJMoa1102673 PMID:21526923 > Crossref MedlineGoogle Scholar - 12.. Cost-effectiveness of ranibizumab and bevacizumab for age-related macular degeneration: 2-year findings from the IVAN randomised trial. BMJ Open. 2014; 4(7):e005094–e005094.
10.1136/bmjopen-2014-005094 PMID:25079928 > Crossref MedlineGoogle Scholar - 13.. A systematic review of the efficacy and safety outcomes of anti-VEGF agents used for treating neovascular age-related macular degeneration: comparison of ranibizumab and bevacizumab. Curr Med Res Opin. 2011; 27(7):1465–1475.
10.1185/03007995.2011.585394 PMID:21623685 > Crossref MedlineGoogle Scholar - 14.. ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology. 2012; 119(7):1388–1398.
10.1016/j.ophtha.2012.03.053 PMID:22555112 > Crossref MedlineGoogle Scholar - 15.. Alternative treatments to inhibit VEGF in age-related choroidal neovascularisation: 2-year findings of the IVAN randomised controlled trial. Lancet. 2013; 382(9900):1258–1267.
10.1016/S0140-6736(13)61501-9 PMID:23870813 > Crossref MedlineGoogle Scholar - 16.. Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial. Ophthalmology. 2012; 119(7):1399–1411.
10.1016/j.ophtha.2012.04.015 PMID:22578446 > Crossref MedlineGoogle Scholar - 17.. Evaluation of compounded bevacizumab prepared for intravitreal injection. JAMA Ophthalmol. 2015; 133(1):32–39.
10.1001/jamaophthalmol.2014.3591 PMID:25233052 > Crossref MedlineGoogle Scholar - 18.. Avastin doesn't blind people, people blind people. Am J Ophthalmol. 2012; 153(2):196–203.e1.
10.1016/j.ajo.2011.11.023 PMID:22264942 > Crossref MedlineGoogle Scholar - 19.. A cluster of presumed, noninfectious endophthalmitis after intravitreal injection of bevacizumab: long-term follow-up. Digit J Ophthalmol. 2016; 22(2):41–45.
10.5693/djo.01.2016.03.001 PMID:27582674 > Crossref MedlineGoogle Scholar - 20.. Quality of bevacizumab compounded for intravitreal administration. Eye (Lond). 2013; 27(9):1090–1097.
10.1038/eye.2013.139 PMID:23788205 > Crossref MedlineGoogle Scholar - 21.. Randomized, double-masked, sham-controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER study year 2. Am J Ophthalmol. 2010; 150(3):315–324.e1.
10.1016/j.ajo.2010.04.011 PMID:20598667 > Crossref MedlineGoogle Scholar - 22.. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011; 364(20):1897–1908.
10.1056/NEJMoa1102673 PMID:21526923 > Crossref MedlineGoogle Scholar - 23.. Intravitreal bevacizumab versus ranibizumab for treatment of neovascular age-related macular degeneration: findings from a Cochrane systematic review. Ophthalmology. 2016; 123(1):70–77.e1.
10.1016/j.ophtha.2015.09.002 PMID:26477843 > Crossref MedlineGoogle Scholar - 24.. Comparing the effectiveness of bevacizumab to ranibizumab in patients with exudative age-related macular degeneration. The BRAMD Study. PLoS One. 2016; 11(5):e0153052.
10.1371/journal.pone.0153052 PMID:27203434 > Crossref MedlineGoogle Scholar - 25.. Efficacy and safety of antivascular endothelial growth factor (anti-VEGF) in treating neovascular age-related macular degeneration (AMD): a systematic review and meta-analysis. J Immunol Res. 2022;(1):6004047.
10.1155/2022/6004047 . > Crossref MedlineGoogle Scholar - 26.. Efficacy and safety of anti-vascular endothelial growth factor monotherapies for neovascular age-related macular degeneration: a mixed treatment comparison. Front Pharmacol. 2021; 12:797108.
10.3389/fphar.2021.797108 PMID:34992542 > Crossref MedlineGoogle Scholar - 27.. Multi-country real-life experience of anti-vascular endothelial growth factor therapy for wet age-related macular degeneration. Br J Ophthalmol. 2015; 99(2):220–226.
10.1136/bjophthalmol-2014-305327 PMID:25193672 > Crossref MedlineGoogle Scholar - 28.. Changes in visual acuity in patients with wet age-related macular degeneration treated with intravitreal ranibizumab in daily clinical practice: the LUMIERE study. Retina. 2013; 33(3):474–481.
10.1097/IAE.0b013e31827b6324 PMID:23266880 > Crossref MedlineGoogle Scholar - 29.. Real-world effects of anti-vascular endothelial growth factor injection frequency on visual outcomes in patients with diabetic macular oedema. Eye (Lond). 2024; 38(9):1687–1693.
10.1038/s41433-024-02998-2 PMID:38448732 > Crossref MedlineGoogle Scholar - 30.Lytenava, inn-bevacizumab gamma. Appendix I–III. European Medicines Agency; 2024. https://ec.europa.eu/health/documents/community-register/2024/20240527162457/anx_162457_en.pdf > Google Scholar
- 31.. Silicone oil microdroplets and protein aggregates in repackaged bevacizumab and ranibizumab: effects of long-term storage and product mishandling. Invest Ophthalmol Vis Sci. 2011; 52(2):1023–1034.
10.1167/iovs.10-6431 PMID:21051703 > Crossref MedlineGoogle Scholar - 32.. High-molecular-weight aggregates in repackaged bevacizumab. Retina. 2010; 30(6):887–892.
10.1097/IAE.0b013e3181d50cea PMID:20458261 > Crossref MedlineGoogle Scholar - 33.. Six-month stability of bevacizumab (Avastin) binding to vascular endothelial growth factor after withdrawal into a syringe and refrigeration or freezing. Retina. 2006; 26(5):519–522.
10.1097/01.iae.0000225354.92444.7a PMID:16770257 > Crossref MedlineGoogle Scholar - 34.. Serratia marcescens endophthalmitis associated with intravitreal injections of bevacizumab. Eye (Lond). 2010; 24(2):226–232.
10.1038/eye.2009.86 PMID:19407833 > Crossref MedlineGoogle Scholar - 35.. Sterile endophthalmitis after intravitreal injection of bevacizumab obtained from a single batch. Retina. 2010; 30(3):485–490.
10.1097/IAE.0b013e3181bd2d51 PMID:19952993 > Crossref MedlineGoogle Scholar - 36.. Severe intraocular inflammation after intravitreal injection of bevacizumab. Ophthalmology. 2010; 117(3):512–516, 516.e1–516.e2.
10.1016/j.ophtha.2009.07.041 PMID:20031229 > Crossref MedlineGoogle Scholar - 37.. Intravitreal bevacizumab. Indian J Ophthalmol. 2018; 66(1):173–174.
10.4103/ijo.IJO_802_17 PMID:29283157 > Crossref MedlineGoogle Scholar - 38.. Acute intraocular inflammation following intravitreal injection of bevacizumab--a large cluster of cases. Acta Ophthalmol. 2011; 89(8):e664–e665.
10.1111/j.1755-3768.2010.02054.x PMID:21159147 > Crossref MedlineGoogle Scholar - 39.. Mechanisms of sterile inflammation after intravitreal injection of antiangiogenic drugs: a narrative review. Int J Retina Vitreous. 2021; 7(1):37.
10.1186/s40942-021-00307-7 PMID:33962696 > Crossref MedlineGoogle Scholar - 40.. Demonstrating substantial evidence of effectiveness for human drug and biological products - guidance for industry. 2019. Accessed June 26, 2023. https://www.fda.gov/media/133660/download > Google Scholar
