YESAFILI Solution for injection Ref.[51275] Active ingredients: Aflibercept

Pharmacotherapeutic group: Ophthalmologicals / Antineovascularisation agents
ATC code: S01LA05

Aflibercept is a recombinant fusion protein consisting of portions of human VEGF receptor 1 and 2 extracellular domains fused to the Fc portion of human IgG1.

Aflibercept is produced in Chinese hamster ovary (CHO) K1 cells by recombinant DNA technology.

Aflibercept acts as a soluble decoy receptor that binds VEGF-A and PlGF with higher affinity than their natural receptors, and thereby can inhibit the binding and activation of these cognate VEGF receptors.

Mechanism of action

Vascular endothelial growth factor-A (VEGF-A) and placental growth factor (PlGF) are members of the VEGF family of angiogenic factors that can act as potent mitogenic, chemotactic, and vascular permeability factors for endothelial cells. VEGF acts via two receptor tyrosine kinases; VEGFR-1 and VEGFR-2, present on the surface of endothelial cells. PlGF binds only to VEGFR-1, which is also present on the surface of leucocytes. Excessive activation of these receptors by VEGF-A can result in pathological neovascularisation and excessive vascular permeability. PlGF can synergize with VEGFA in these processes and is also known to promote leucocyte infiltration and vascular inflammation.

Pharmacodynamic effects

wet AMD

Wet AMD is characterised by pathological choroidal neovascularisation (CNV). Leakage of blood and fluid from CNV may cause retinal thickening or oedema and/or sub-/intra-retinal haemorrhage, resulting in loss of visual acuity.

In patients treated with aflibercept (one injection per month for three consecutive months, followed by one injection every 2 months), central retinal thickness [CRT] decreased soon after treatment initiation, and the mean CNV lesion size was reduced, consistent with the results seen with ranibizumab 0.5 mg every month.

In the VIEW1 study there were mean decreases in CRT on optical coherence tomography (OCT) (-130 and -129 microns at week 52 for the aflibercept 2 mg every two months and ranibizumab 0.5 mg every month study groups, respectively). Also at the 52-week time point, in the VIEW2 study there were mean decreases in CRT on OCT (-149 and -139 microns for the aflibercept 2 mg every two months and ranibizumab 0.5 mg every month study groups, respectively). The reduction of CNV size and reduction in CRT were generally maintained in the second year of the studies.

The ALTAIR study was conducted in Japanese patients with treatment naïve wet AMD, showing similar outcomes to the VIEW studies using 3 initial monthly aflibercept 2 mg injections, followed by one injection after a further 2 months, and then continued with a treat-and-extend regimen with variable treatment intervals (2-week or 4-week adjustments) up to a maximum 16 week interval according to pre-specified criteria. At week 52, there were mean decreases in central retinal thickness (CRT) on OCT of -134.4 and -126.1 microns for the 2-week adjustment group and the 4-week adjustment group, respectively. The proportion of patients without fluid on OCT at week 52 was 68.3% and 69.1% in the 2- and 4-week adjustment groups, respectively. The reduction in CRT was generally maintained in both treatment arms in the second year of the ALTAIR study.

The ARIES study was designed to explore the non-inferiority of an aflibercept 2 mg treat-and-extend dosing regimen initiated immediately after administration of 3 initial monthly injections and one additional injection after 2 months vs. a treat-and-extend dosing regimen initiated after one year of treatment. For patients requiring a more frequent than Q8 dosing at least once over the course of the study, CRT remained higher, but the mean decrease in CRT from baseline to week 104 was -160.4 microns, similar to the patients treated at Q8 or less frequent intervals.

Macular oedema secondary to CRVO and BRVO

In CRVO and BRVO, retinal ischaemia occurs and signals the release of VEGF which in turn destabilises the tight junctions, and promotes endothelial cell proliferation. Up-regulation of VEGF is associated with the breakdown of the blood retina barrier, increased vascular permeability, retinal oedema, and neovascularisation complications.

In patients treated with 6 consecutive monthly injections of aflibercept 2 mg, there was a consistent, rapid and robust morphologic response (as measured by improvements in mean CRT) observed. At week 24, the reduction in CRT was statistically superior versus control in all three studies (COPERNICUS in CRVO: -457 vs. -145 microns; GALILEO in CRVO: -449 vs. -169 microns; VIBRANT in BRVO: -280 vs. -128 microns). This decrease from baseline in CRT was maintained to the end of each study, week 100 in COPERNICUS, week 76 in GALILEO, and week 52 in VIBRANT.

Diabetic macular oedema

In patients treated with aflibercept, the majority of whom were classified as having Type II diabetes, a rapid and robust response in morphology (CRT, DRSS level) was observed.

In the VIVID^DME and the VISTA^DME studies, a statistically significant greater mean decrease in CRT from baseline to week 52 was observed in patients treated with aflibercept than with the laser control, -192.4 and -183.1 microns for the 2Q8 aflibercept groups and -66.2 and -73.3 microns for the control groups, respectively. At week 100 the decrease was maintained with -195.8 and -191.1 microns for the 2Q8 aflibercept groups and -85.7 and -83.9 microns for the control groups, in the VIVID^DME and VISTA^DME studies, respectively.

A ≥2 step improvement in DRSS was assessed in a pre-specified manner in VIVID^DME and VISTA^DME. The DRSS score was gradable in 73.7% of the patients in VIVID^DME and 98.3% of the patients in VISTA^DME. At week 52, 27.7% and 29.1% of the aflibercept 2Q8 groups, and 7.5% and 14.3% of the control groups experienced a ≥2 step improvement in the DRSS. At week 100, the respective percentages were 32.6% and 37.1% of the aflibercept 2Q8 groups and 8.2% and 15.6% of the control groups.

The VIOLET study compared three different dosing regimens of aflibercept 2 mg for treatment of DME after at least one year of treatment at fixed intervals, where treatment was initiated with 5 consecutive monthly doses followed by dosing every 2 months. At week 52 and week 100 of the study, i.e. second and third year of treatment, the mean changes in CRT were clinically similar for treat-and-extend (2T&E), pro re nata (2PRN) and 2Q8, respectively, -2.1, 2.2 and -18.8 microns at week 52, and 2.3, -13.9 and -15.5 microns at week 100.

Myopic choroidal neovascularisation

Myopic choroidal neovascularisation (myopic CNV) is a frequent cause of vision loss in adults with pathologic myopia. It develops as a wound healing mechanism consequent to Bruch’s membrane ruptures and represents the most vision-threatening event in pathologic myopia.

In patients treated with aflibercept in the MYRROR study (one injection given at start of therapy, with additional injections given in case of disease persistence or recurrence), CRT decreased soon after treatment initiation favouring aflibercept at week 24 (-79 microns and -4 microns for the aflibercept 2 mg treatment group and the control group, respectively), which was maintained through week 48. In addition, the mean CNV lesion size decreased.

Clinical efficacy and safety

wet AMD

The safety and efficacy of aflibercept were assessed in two randomised, multi-centre, double-masked, active-controlled studies in patients with wet AMD (VIEW1 and VIEW2) with a. total of 2 412 patients treated and evaluable for efficacy (1 817 with aflibercept). Patient ages ranged from 49 to 99 years with a mean of 76 years. In these clinical studies, approximately 89% (1 616/1 817) of the patients randomised to treatment with aflibercept were 65 years of age or older, and approximately 63% (1 139/1 817) were 75 years of age or older. In each study, patients were randomly assigned in a 1:1:1:1 ratio to 1 of 4 dosing regimens:

1) Aflibercept administered at 2 mg every 8 weeks following 3 initial monthly doses (aflibercept 2Q8);

2) Aflibercept administered at 2 mg every 4 weeks (aflibercept 2Q4);

3) Aflibercept administered at 0.5 mg every 4 weeks (aflibercept 0.5Q4); and

4) ranibizumab administered at 0.5 mg every 4 weeks (ranibizumab 0.5Q4).

In the second year of the studies, patients continued to receive the initially randomised dose but on a modified dosing schedule guided by assessment of visual and anatomic outcomes with a protocoldefined maximum dosing interval of 12 weeks.

In both studies, the primary efficacy endpoint was the proportion of patients in the Per Protocol Set who maintained vision, i.e. losing fewer than 15 letters of visual acuity at week 52 from baseline.

In the VIEW1 study, at week 52, 95.1% of patients in the aflibercept 2Q8 group maintained vision compared to 94.4% patients in the ranibizumab 0.5Q4 group. In the VIEW2 study, at week 52, 95.6% of patients in the aflibercept 2Q8 group maintained vision compared to 94.4% patients in the ranibizumab 0.5Q4 group. In both studies aflibercept was shown to be non-inferior and clinically equivalent to the ranibizumab 0.5Q4 group.

Detailed results from the combined analysis of both studies are shown in Table 2 and Figure 1 below.

Table 2. Efficacy outcomes at week 52 (primary analysis) and week 96; combined data from the VIEW1 and VIEW2 studies^B:

^A BCVA: Best Corrected Visual Acuity
ETDRS: Early Treatment Diabetic Retinopathy Study
LS: Least square means derived from ANCOVA
PPS: Per Protocol Set
^B Full Analysis Set (FAS), Last Observation Carried Forward (LOCF) for all analyses except proportion of patients with maintained visual acuity at week 52 which is PPS
^C The difference is the value of the aflibercept group minus the value of the ranibizumab group. A positive value favours aflibercept
^D Confidence interval (CI) calculated by normal approximation
^E After treatment initiation with three monthly doses
^F A confidence intervallying entirely above -10% indicates a non-inferiority of aflibercept to ranibizumab

Figure 1. Mean change in visual acuity from baseline to week 96 for the combined data from the view 1 and view 2 studies:

In combined data analysis of VIEW1 and VIEW2, aflibercept demonstrated clinically meaningful changes from baseline in pre-specified secondary efficacy endpoint National Eye Institute Visual Function Questionnaire (NEI VFQ-25) without clinically meaningful differences to ranibizumab. The magnitude of these changes was similar to that seen in published studies, which corresponded to a 15-letter gain in Best Corrected Visual Acuity (BCVA).

In the second year of the studies, efficacy was generally maintained through the last assessment at week 96, and 2-4% of patients required all injections on a monthly basis, and a third of patients required at least one injection with a treatment interval of only one month.

Decreases in mean CNV area were evident in all dose groups in both studies.

Efficacy results in all evaluable subgroups (e.g., age, gender, race, baseline visual acuity, lesion type, lesion size) in each study and in the combined analysis were consistent with the results in the overall populations.

ALTAIR was a 96-week multicentre, randomised, open-label study in 247 Japanese patients with treatment naïve wet AMD, designed to assess the efficacy and safety of aflibercept following two different adjustment intervals (2-weeks and 4-weeks) of a treat-and-extend dosing regimen.

All patients received monthly doses of aflibercept 2 mg for 3 months, followed by one injection after a further 2-month interval. At week 16, patients were randomised 1:1 into two treatment groups: 1) aflibercept treat-and-extend with 2-week adjustments and 2) aflibercept treat-and-extend with 4 -week adjustments. Extension or shortening of the treatment interval was decided based on visual and/or anatomic criteria defined by protocol with a maximum treatment interval of 16 weeks for both groups.

The primary efficacy endpoint was mean change in BCVA from baseline to week 52. The secondary efficacy endpoints were the proportion of patients who did not lose ≥ 15 letters and the proportion of patients who gained at least 15 letters of BCVA from baseline to week 52.

At week 52, patients in the treat-and-extend arm with 2 -week adjustments gained a mean of 9.0 letters from baseline as compared to 8.4 letters for those in the 4-week adjustment group [LS mean difference in letters (95% CI): -0.4 (-3.8,3.0), ANCOVA]. The proportion of patients who did not lose ≥15 letters in the two treatment arms was similar (96.7% in the 2-week and 95.9% in the 4-week adjustment groups). The proportion of patients who gained ≥15 letters at week 52 was 32.5% in the 2-week adjustment group and 30.9% in the 4-week adjustment group. The proportion of patients who extended their treatment interval to 12 weeks or beyond was 42.3% in the 2-week adjustment group and 49.6% in the 4-week adjustment group. Furthermore, in the 4 -week adjustment group 40.7% of patients were extended to 16-week intervals. At the last visit up to week 52, 56.8% and 57.8% of patients in the 2 -week and 4 -week adjustment groups, respectively had their next injection scheduled at an interval of 12 weeks or beyond.

In the second year of the study, efficacy was generally maintained up to and including the last assessment at week 96, with a mean gain from baseline of 7.6 letters for the 2-week adjustment group and 6.1 letters for the 4 -week adjustment group. The proportion of patients who extended their treatment interval to 12 weeks or beyond was 56.9% in the 2 -week adjustment group and 60.2% in the 4-week adjustment group. At the last visit prior to week 96, 64.9% and 61.2% of patients in the 2-week and 4-week adjustment groups, respectively had their next injection scheduled at an interval of 12 weeks or beyond. During the second year of treatment patients in both the 2-week and 4-week adjustment groups received an average of 3.6 and 3.7 injections, respectively. Over the 2-year treatment period patients received an average of 10.4 injections.

Ocular and systemic safety profiles were similar to the safety observed in the pivotal studies VIEW1 and VIEW2.

ARIES was a 104 -week multicentre, randomised, open-label, active-controlled study in 269 patients with treatment naïve wet AMD, designed to assess the non-inferiority in terms of efficacy as well as the safety of a treat-and-extend dosing regimen initiated after 3 consecutive monthly doses followed by extension to a 2 monthly treatment interval vs. a treat-and-extend dosing regimen initiated after the first year of treatment.

The ARIES study also explored the percentage of patients that required more frequent treatment than every 8 weeks based on the investigator’s decision. Out of the 269 patients 62 patients received more frequent dosing at least once during the course of the study. Such patients remained in the study and received treatment according to the investigator’s best clinical judgement but not more frequently than every 4 weeks and their treatment intervals could be extended again afterwards. The average treatment interval after the decision to treat more frequently was 6.1 weeks. Week 104 BCVA was lower in patients requiring more intensive treatment at least once over the course of the study compared with patients who did not and the mean change in BCVA from baseline to end of the study was + 2.3 ± 15.6 letters. Among the patients treated more frequently, 85.5% maintained vision, i.e., lost less than15 letters, and 19.4% gained 15 letters or more. The safety profile of patients treated more frequently than every 8 weeks was comparable to the safety data in VIEW 1 and VIEW 2.

Macular oedema secondary to CRVO

The safety and efficacy of aflibercept were assessed in two randomised, multi-centre, double-masked, sham-controlled studies in patients with macular oedema secondary to CRVO (COPERNICUS and GALILEO) with a total of 358 patients treated and evaluable for efficacy (217 with aflibercept). Patient ages ranged from 22 to 89 years with a mean of 64 years. In the CRVO studies, approximately 52% (112/217) of the patients randomised to treatment with aflibercept were 65 years of age or older, and approximately 18% (38/217) were 75 years of age or older. In both studies, patients were randomly assigned in a 3:2 ratio to either 2 mg aflibercept administered every 4 weeks (2Q4), or the control group receiving sham injections every 4 weeks for a total of 6 injections.

After 6 consecutive monthly injections, patients received treatment only if they met pre-specified retreatment criteria, except for patients in the control group in the GALILEO study who continued to receive sham (control to control) until week 52. From this timepoint all patients were treated if prespecified criteria were met.

In both studies, the primary efficacy endpoint was the proportion of patients who gained at least 15 letters in BCVA at week 24 compared to baseline. A secondary efficacy variable was change in visual acuity at week 24 compared to baseline.

The difference between treatment groups was statistically significant in favour of aflibercept in both studies. The maximal improvement in visual acuity was achieved at month 3 with subsequent stabilisation of visual acuity and CRT until month 6. The statistically significant difference was maintained through week 52.

Detailed results from the analysis of both studies are shown in Table 3 and Figure 2 below.

Table 3. Efficacy outcomes at week 24, week 52 and week 76/100 (full analysis set with LOCF^C)) in COPERNICUS and GALILEO studies:

^A Difference is aflibercept 2 mg Q4 weeks minus control
^B Difference and confidence interval (CI) are calculated using Cochran-Mantel-Haenszel (CMH) test adjusted for region (America vs. rest of the world for COPERNICUS and Europe vs. Asia/Pacific for GALILEO) and baseline BCVA category (>20/200 and ≤20/200)
^C BCVA: Best Corrected Visual Acuity ETDRS: Early Treatment Diabetic Retinopathy Study LOCF: Last Observation Carried Forward SD: Standard deviation LS: Least square means derived from ANCOVA
^D LS mean difference and confidence interval based on an ANCOVA model with factors treatment group, region (America vs. rest of the world for COPERNICUS and Europe vs. Asia/Pacific for GALILEO) and baseline BCVA category (>20/200 and ≤20/200)
^E In COPERNICUS study, control group patients could receive aflibercept on an as-needed basis as frequently as every 4 weeks during week 24 to week 52; patients had visits every 4 weeks.
^F In COPERNICUS study, both control group and aflibercept 2 mg patients received aflibercept 2 mg on an as-needed basis as frequently as every 4 weeks starting from week 52 to week 96; patients had mandatory quarterly visits but may have been seen as frequently as every 4 weeks if necessary.
^G In GALILEO study, both control group and aflibercept 2 mg patients received aflibercept 2 mg on an as-needed basis every 8 weeks starting from week 52 to week 68; patients had mandatory visits every 8 weeks.

Figure 2. Mean change from baseline to week 76/100 in visual acuity by treatment group for the COPERNICUS and GALILEO studies (full analysis set):

In GALILEO, 86.4% (n=89) of the aflibercept group and 79.4% (n=54) of the sham group had perfused CRVO at baseline. At week 24, this was 91.8% (n=89) in the aflibercept group and 85.5% (n=47) in the sham group. These proportions were maintained at week 76, with 84.3% (n=75) in the aflibercept group and 84.0% (n=42) in the sham group.

In COPERNICUS, 67.5% (n=77) of the aflibercept group and 68.5% (n=50) of the sham group had perfused CRVO at baseline. At week 24, this was 87.4% (n=90) in the aflibercept group and 58.6% (n=34) in the sham group. These proportions were maintained at week 100 with 76.8% (n=76) in the aflibercept group and 78% (n=39) in the sham group. Patients in the sham group were eligible to receive aflibercept from week 24.

The beneficial effect of aflibercept treatment on visual function was similar in the baseline subgroups of perfused and non-perfused patients. Treatment effects in other evaluable subgroups (e.g., age, gender, race, baseline visual acuity, CRVO duration) in each study were in general consistent with the results in the overall populations.

In combined data analysis of GALILEO and COPERNICUS, aflibercept demonstrated clinically meaningful changes from baseline in pre-specified secondary efficacy endpoint National Eye Institute Visual Function Questionnaire (NEI VFQ-25). The magnitude of these changes was similar to that seen in published studies, which corresponded to a 15-letter gain in Best Corrected Visual Acuity (BCVA).

Macular oedema secondary to BRVO

The safety and efficacy of aflibercept were assessed in a randomised, multi-centre, double-masked, active-controlled study in patients with macular oedema secondary to BRVO (VIBRANT) which included Hemi-Retinal Vein Occlusion. A total of 181 patients were treated and evaluable for efficacy (91 with aflibercept). Patient ages ranged from 42 to 94 years with a mean of 65 years. In the BRVO study, approximately 58% (53/91) of the patients randomised to treatment with aflibercept were 65 years of age or older, and approximately 23% (21/91) were 75 years of age or older. In the study, patients were randomly assigned in a 1:1 ratio to either 2 mg aflibercept administered every 8 weeks following 6 initial monthly injections or laser photocoagulation administered at baseline (laser control group). Patients in the laser control group could receive additional laser photocoagulation (called ‘rescue laser treatment’) beginning at week 12 with a minimum interval of 12 weeks. Based on prespecified criteria, patients in the laser group could receive rescue treatment with aflibercept 2 mg from week 24, administered every 4 weeks for 3 months followed by every 8 weeks.

In the VIBRANT study, the primary efficacy endpoint was the proportion of patients who gained at least 15 letters in BCVA at week 24 compared to baseline and the aflibercept group was superior to laser control.

A secondary efficacy endpoint was change in visual acuity at week 24 compared to baseline, which was statistically significant in favour of aflibercept in the VIBRANT study. The course of visual improvement was rapid and peaked at 3 months with maintenance of the effect until month 12.

In the laser group 67 patients received rescue treatment with aflibercept beginning at week 24 (Active Control/ aflibercept 2 mg group), which resulted in improvement of visual acuity by about 5 letters from week 24 to 52.

Detailed results from the analysis of the VIBRANT study are shown in Table 4 and Figure 3 below.

Table 4. Efficacy outcomes at week 24 and week 52 (full analysis set with LOCF) in VIBRANT study:

^A Difference is aflibercept 2 mg Q4 weeks minus Laser Control
^B Difference and 95% CI are calculated using Mantel-Haenszel weighting scheme adjusted for region (North America vs. Japan) and baseline BCVA category (>20/200 and ≤20/200)
^C LS mean difference and 95% CI based on an ANCOVA model with treatment group, baseline BCVA category (>20/200 and ≤20/200) and region (North America vs. Japan) as fixed effects, and baseline BCVA as covariate.
^D From week 24 on the treatment interval in the aflibercept treatment group was extended for all subjects from 4 weeks to 8 weeks through week 48.
^E Beginning at week 24 subjects in the Laser Group could receive rescue treatment with aflibercept, if they met at least one pre-specified eligibility criterion. At total of 67 subjects in this group received aflibercept rescue treatment. The fixed regimen for aflibercept rescue was three times aflibercept 2 mg every 4 weeks followed by injections every 8 weeks.
^F Nominal p-value

Figure 3. Mean change in BCVA as measured by ETDRS letter score from baseline to week 52 in VIBRANT Study:

At baseline, the proportion of perfused patients in the aflibercept and laser groups was 60% and 68%, respectively. At week 24 these proportions were 80% and 67%, respectively. In the aflibercept group the proportion of perfused patients was maintained through week 52. In the laser group, where patients were eligible for rescue treatment with aflibercept from week 24, the proportion of perfused patients increased to 78% by week 52.

Diabetic macular oedema

The safety and efficacy of aflibercept were assessed in two randomised, multi-centre, double-masked, active-controlled studies in patients with DME (VIVID^DME and VISTA^DME). A total of 862 patients were treated and evaluable for efficacy, 576 with aflibercept. Patient ages ranged from 23 to 87 years with a mean of 63 years. In the DME studies, approximately 47% (268/576) of the patients randomised to treatment with aflibercept were 65 years of age or older, and approximately 9% (52/576) were 75 years of age or older. The majority of patients in both studies had Type II diabetes.

In both studies, patients were randomly assigned in a 1:1:1 ratio to 1 of 3 dosing regimens:

1) Aflibercept administered 2 mg every 8 weeks following 5 initial monthly injections (aflibercept 2Q8);

2) Aflibercept administered 2 mg every 4 weeks (aflibercept 2Q4); and

3) Macular laser photocoagulation (active control).

Beginning at week 24, patients meeting a pre-specified threshold of vision loss were eligible to receive additional treatment: patients in the aflibercept groups could receive laser and patients in the control group could receive aflibercept.

In both studies, the primary efficacy endpoint was the mean change from baseline in BCVA at week 52 and both aflibercept 2Q8 and aflibercept 2Q4 groups demonstrated statistical significance and were superior to the control group. This benefit was maintained through week 100.

Detailed results from the analysis of the VIVID^DME and VISTA^DME studies are shown in Table 5 and Figure 4 below.

Table 5. Efficacy outcomes at week 52 and week 100 (full analysis set with LOCF) in VIVID^DME and VISTA^DME studies:

^A After treatment initiation with 5 monthly injections
^B LS mean and CI based on an ANCOVA model with baseline BCVA measurement as a covariate and a factor for treatment group. Additionally, region (Europe/Australia vs. Japan) had been included as factor for VIVIDDME, and history of MI and/or CVA as a factor for VISTADME
^C Difference is aflibercept group minus active control (laser) group
^D Difference with confidence interval (CI) and statistical test is calculated using Mantel-Haenszel weighting scheme adjusted by region (Europe/Australia vs. Japan) for VIVIDDME and medical history of MI or CVA for VISTADME
^E BCVA: Best Corrected Visual Acuity
ETDRS: Early Treatment Diabetic Retinopathy Study; LOCF: Last Observation Carried Forward; LS: Least square means derived from ANCOVA; CI: Confidence interval

Figure 4. Mean change in BCVA as measured by ETDRS letter score from baseline to week 100 in VIVID^DME and VISTA^DME studies:

Treatment effects in evaluable subgroups (e.g., age, gender, race, baseline HbA1c, baseline visual acuity, prior anti-VEGF therapy) in each study and in the combined analysis were generally consistent with the results in the overall populations.

In the VIVID^DME and VISTA^DME studies, 36 (9%) and 197 (43%) patients received prior anti-VEGF therapy, respectively, with a 3-month or longer washout period. Treatment effects in the subgroup of patients who had previously been treated with a VEGF inhibitor were similar to those seen in patients who were VEGF inhibitor naïve.

Patients with bilateral disease were eligible to receive anti-VEGF treatment in their fellow eye if assessed necessary by the physician. In the VISTA^DME study, 217 (70.7%) of aflibercept patients received bilateral aflibercept injections until week100; in the VIVID^DME study, 97 (35.8%) of aflibercept patients received a different anti-VEGF treatment in their fellow eye.

An independent comparative trial (DRCR.net Protocol T) utilised a flexible dosing regimen based on strict OCT and vision re-treatment criteria. In the aflibercept treatment group (n=224) at week 52, this treatment regimen resulted in patients receiving a mean of 9.2 injections, which is similar to the administered number of doses in the aflibercept 2Q8 group in VIVID^DME and VISTA^DME, while overall efficacy of the aflibercept treatment group in Protocol T was comparable to the aflibercept 2Q8 group in VIVID^DME and VISTA^DME. A 13.3 mean letter gain with 42% of patients gaining at least 15 letters in vision from baseline was observed in Protocol T. Safety outcomes demonstrated that overall incidences of ocular and non-ocular adverse events (including ATEs) were comparable across all treatment groups in each of the studies and between the studies.

VIOLET, a 100-week multicentre, randomised, open-label, active controlled study in patients with DME compared three different dosing regimens of aflibercept 2 mg for treatment of DME after at least one year of treatment at fixed intervals, where treatment was initiated with 5 consecutive monthly doses followed by dosing every 2 months. The study evaluated non-inferiority of aflibercept 2 mg dosed according to a treat-and-extend regimen (2T&E where injections intervals were kept at a minimum of 8 weeks and gradually extended based on clinical and anatomical outcomes) and aflibercept 2 mg dosed as needed (2PRN where patients were observed every 4 weeks and injected when needed based on clinical and anatomical outcomes), compared to aflibercept 2 mg dosed every 8 weeks (2Q8) for the second and third year of treatment.

The primary efficacy endpoint (change in BCVA from baseline to week 52) was 0.5 ± 6.7 letters in the 2T&E group and 1.7 ± 6.8 letters in the 2PRN group compared to 0.4 ± 6.7 letters in the 2Q8 group, achieving statistical non-inferiority (p<0.0001 for both comparisons; NI margin 4 letters). The changes in BCVA from baseline to week 100 were consistent with the week 52 results: -0.1 ± 9.1 letters in the 2T&E group and 1.8 ± 9.0 letters in the 2PRN group compared to 0.1 ± 7.2 letters in the 2Q8 group. The mean number of injections over 100 weeks were 12.3, 10.0 and 11.5 for 2Q8fix, 2T&E and 2PRN, respectively.

Ocular and systemic safety profiles in all 3 treatment groups were similar to those observed in the pivotal studies VIVID and VISTA.

In the 2T&E group, the increments and decrements for the injection intervals were at the investigator’s discretion; increments of 2 weeks were recommended in the study.

Myopic choroidal neovascularisation

The safety and efficacy of aflibercept were assessed in a randomised, multi-centre, double-masked, sham-controlled study in treatment-naïve, Asian patients with myopic CNV. A total of 121 patients were treated and evaluable for efficacy (90 with aflibercept). Patient ages ranged from 27 to 83 years with a mean of 58 years. In the myopic CNV study, approximately 36% (33/91) of the patients randomised to treatment with aflibercept were 65 years of age or older, and approximately 10% (9/91) were 75 years of age or older.

Patients were randomly assigned in a 3:1 ratio to receive either 2 mg aflibercept intravitreally or sham injections administered once at study start with additional injections given monthly in case of disease persistence or recurrence until week 24, when the primary endpoint was assessed. At week 24, patients initially randomised to sham were eligible to receive the first dose of aflibercept. Following this, patients in both groups continued to be eligible for additional injections in case of disease persistence or recurrence.

The difference between treatment groups was statistically significant in favour of aflibercept for the primary endpoint (change in BCVA) and confirmatory secondary efficacy endpoint (proportion of patients who gained 15 letters in BCVA) at week 24 compared to baseline. Differences for both endpoints were maintained through week 48.

Detailed results from the analysis of the MYRROR study are shown in Table 6 and Figure 5 below.

Table 6. Efficacy outcomes at week 24 (primary analysis) and week 48 in MYRROR study (full analysis set with LOCF^A)):

^A LOCF: Last Observation Carried Forward
^B BCVA: Best Corrected Visual Acuity
ETDRS: Early Treatment Diabetic Retinopathy Study
SD: Standard Deviation
^C LS mean: Least square means derived from ANCOVA model
^D CI: Confidence Interval
^E LS mean difference and 95% CI based on an ANCOVA model with treatment group and country (country designations) as fixed effects, and baseline BCVA as covariant.
^F Difference and 95% CI are calculated using Cochran-Mantel-Haenszel (CMH) test adjusted for country (country designations)

Figure 5. Mean change from baseline to week 48 in visual acuity by treatment group for the MYRROR study (full analysis set, LOCF):

Paediatric population

The European Medicines Agency has waived the obligation to submit the results of studies with aflibercept in all subsets of the paediatric population in wet AMD, CRVO, BRVO, DME and myopic CNV populations (see section 4.2 for information on paediatric use).