Clinical Outcomes of Aflibercept in AMD and DME

What clinical outcomes support aflibercept treatment of wet AMD and diabetic macular edema?

Aflibercept treatment for wet AMD and diabetic macular edema is supported by visual acuity improvements equivalent to ranibizumab and superior to laser photocoagulation, significant anatomical improvements including central retinal thickness reductions and diabetic retinopathy severity improvements, clinically meaningful quality of life gains, and a safety profile comparable to other anti-VEGF agents with the advantage of less frequent dosing.

Abstract

Aflibercept treatment for wet age-related macular degeneration produces visual acuity outcomes equivalent to ranibizumab, with both treatments showing similar mean BCVA changes (within 0.5 letters) and approximately 32% of patients gaining ≥15 letters at one year. Anatomical outcomes were also similar, with comparable proportions achieving dry retina (RR 1.06, 95% CI 0.98 to 1.14) and equivalent choroidal neovascularization area reductions. For diabetic macular edema, aflibercept demonstrated statistically significant superiority over laser photocoagulation, with mean BCVA gains of 10.5-13.6 letters compared to -0.5 to 1.2 letters for laser (p<0.0001), and central retinal thickness reductions of 183-195 μm versus 66-73 μm for laser (p<0.0001). At 148 weeks, 35.8-42.9% of aflibercept-treated eyes gained ≥15 letters versus 13.6-18.9% with laser. Aflibercept showed non-inferiority to bevacizumab (mean improvement 15.0 versus 14.0 letters, p=0.37) and better anatomical outcomes than ranibizumab in real-world practice (central subfield thickness reduction -114.4 μm versus -87.8 μm, p<0.01). Quality of life improved meaningfully in both conditions, with NEI VFQ-25 total score gains of +6.11 points in DME and clinically significant improvements across multiple subscales in wet AMD. The safety profile was comparable to other anti-VEGF agents, with serious systemic adverse events similar to ranibizumab (RR 0.99, 95% CI 0.79 to 1.25) and no new safety signals identified in long-term use. The every-8-weeks maintenance dosing regimen after initial loading reduced treatment burden compared to monthly regimens while maintaining efficacy.

Methods

We analyzed 10 sources from an initial pool of 200, using 8 screening criteria. Each paper was reviewed for 8 key aspects that mattered most to the research question.

Records from Elicit search

• n = 200

Papers screened using:

• Target Population, Intervention, Clinical Outcomes, Study Design, Follow-up Duration, Aflibercept Analysis, Study Quality and Accessibility, Study Scope

• n = 200 Papers screened out

• n = 190 Papers included for extraction

Paper search

We performed a semantic search across over 138 million academic papers from the Elicit search engine, which includes all of Semantic Scholar and OpenAlex.

We ran this query: "What clinical outcomes support aflibercept treatment of wet AMD and diabetic macular edema?"

The search returned 200 total results from Elicit.

We retrieved 200 papers most relevant to the query for screening.

Screening

We screened in sources based on their abstracts that met these criteria:

• Target Population: Does the study include patients diagnosed with wet AMD (neovascular age-related macular degeneration) or diabetic macular edema?
• Intervention: Does the study examine aflibercept (Eylea) administered as intravitreal injection?
• Clinical Outcomes: Does the study report clinical outcomes related to visual function, anatomical changes, or safety?
• Study Design: Is the study a randomized controlled trial, controlled clinical trial, large prospective cohort study (n≥50), or systematic review/meta-analysis?
• Follow-up Duration: Does the study have a minimum 12-week follow-up period?
• Aflibercept Analysis: Can aflibercept outcomes be separately analyzed in this study (i.e., not combined with other therapies where aflibercept effects cannot be isolated)?
• Study Quality and Accessibility: Is the study NOT a case report, case series with <10 patients, editorial/commentary, or conference abstract without full-text publication?
• Study Scope: Does the study focus on ophthalmic indications and specifically on wet AMD or diabetic macular edema (rather than other retinal conditions or non-ophthalmic uses)?

We considered all screening questions together and made a holistic judgement about whether to screen in each paper.

Data extraction

We asked a large language model to extract each data column below from each paper.

• Study Design:
  • Study type (RCT, observational, etc.)
  • Sample size and number of eyes treated with aflibercept
  • Patient population (wet AMD or diabetic macular edema or both)
  • Baseline disease characteristics (severity, prior treatment status)
  • Comparator group(s) if any (other anti-VEGF agents, laser, sham, etc.)
  • Follow-up duration
• Visual Outcomes:
  • Best-corrected visual acuity (BCVA) changes from baseline (mean change in ETDRS letters)
  • Proportion gaining ≥15 letters
  • Proportion losing ≥15 letters
  • Proportion achieving specific visual acuity thresholds (e.g., ≥70 letters, 20/40 or better)
  • Low vision outcomes (proportion with vision worse than 20/200)
  • Statistical significance and confidence intervals
  • Time points assessed (52 weeks, 2 years, etc.)
  • Whether outcomes support aflibercept effectiveness
• Anatomical Outcomes:
  • Central retinal thickness changes
  • Proportion achieving dry retina (absence of fluid)
  • Choroidal neovascularization area changes (for wet AMD)
  • Diabetic retinopathy severity scale improvements (for DME)
  • Fluorescein angiography leakage findings
  • Optical coherence tomography findings
  • Statistical significance and effect sizes
  • Whether anatomical improvements support aflibercept treatment
• Quality of Life:
  • Vision-related quality of life scales (NEI VFQ-25 scores)
  • Functional vision assessments (near activities, distance activities)
  • Patient satisfaction measures
  • Vision-related disability assessments
  • Changes from baseline and clinical significance
  • Better-seeing eye vs worse-seeing eye outcomes if reported
  • Whether quality of life improvements support aflibercept use
• Safety Profile:
  • Serious ocular adverse events (endophthalmitis, retinal detachment, cataract, etc.)
  • Serious systemic adverse events
  • Treatment-emergent adverse events by category
  • Rates and risk ratios compared to control groups
  • Deaths and relationship to treatment
  • Injection-related complications
  • Long-term safety findings
  • Whether safety profile supports aflibercept use in these conditions
• Treatment Regimen:
  • Dose administered (typically 2mg)
  • Initial loading schedule (e.g., monthly for 3-5 doses)
  • Maintenance schedule (e.g., every 8 weeks, treat-and-extend)
  • Total number of injections over study period
  • Rescue treatment protocols
  • Treatment burden compared to other regimens
  • Retreatment criteria used
  • Whether dosing regimen is practical and effective
• Comparative Effectiveness:
  • Direct comparisons to ranibizumab, bevacizumab, or laser photocoagulation
  • Non-inferiority or superiority findings
  • Head-to-head efficacy outcomes with effect sizes
  • Real-world effectiveness vs clinical trial results
  • Cost-effectiveness considerations if mentioned
  • Whether comparative data support choosing aflibercept over alternatives
• Key Limitations:
  • Risk of bias concerns (funding source, blinding, etc.)
  • Generalizability limitations (population, setting, exclusion criteria)
  • Missing data or loss to follow-up
  • Protocol deviations affecting aflibercept groups
  • Short follow-up duration for long-term outcomes
  • Confounding factors
  • Any factors that might limit confidence in supporting aflibercept treatment

Results

Characteristics of included studies

The systematic review included 10 sources examining aflibercept treatment for wet age-related macular degeneration (AMD) and diabetic macular edema (DME). These comprised randomized controlled trials, observational studies, and systematic reviews.

The studies represented diverse designs, with five focusing on wet AMD and five on DME. Most studies were randomized controlled trials, with comparators including ranibizumab, bevacizumab, and laser photocoagulation. Follow-up durations ranged from 52 weeks to 3 years. Only two studies had full texts available for detailed extraction (J. Heier et al., 2012 and J. Carrasco et al., 2021).

Effects

Visual Acuity Outcomes in Wet AMD

Aflibercept demonstrated substantial and sustained visual acuity improvements in wet AMD across multiple studies. In the VIEW trials comparing aflibercept to ranibizumab, all aflibercept regimens were within 0.5 letters of ranibizumab for mean change in BCVA, demonstrating clinical equivalence. At one year, the mean change in BCVA was similar between aflibercept and ranibizumab groups (mean difference -0.15 ETDRS letters, 95% CI -1.47 to 1.17). This similarity persisted at two years, with mean changes of 7.2 letters for aflibercept versus 7.9 letters for ranibizumab.

The real-world meta-analysis reported a mean BCVA of 62.55 ETDRS letters at 2 years, with 47.39% of patients achieving visual acuity of ≥70 letters. The mean gain from baseline was +4.49 letters for the combined population and +5.91 letters for patients treated with a treat-and-extend regimen. Vision loss was minimal, with similar small proportions in aflibercept and ranibizumab groups losing ≥15 letters at one year (RR 0.89, 95% CI 0.61 to 1.30).

Visual Acuity Outcomes in Diabetic Macular Edema

Aflibercept demonstrated consistent superiority over laser photocoagulation and non-inferiority to other anti-VEGF agents in DME. In the VISTA and VIVID trials, mean BCVA gains with aflibercept were 12.5 and 10.7 letters versus 0.2 letters with laser in VISTA, and 10.5 and 10.7 letters versus 1.2 letters with laser in VIVID (p<0.0001 for all comparisons).

Long-term outcomes remained robust. At 148 weeks, mean BCVA gains with aflibercept 2q4 and 2q8 were 10.4 and 10.5 letters respectively in VISTA, and 10.3 and 11.7 letters in VIVID, compared to 1.4 and 1.6 letters with laser control (p<0.0001 for all). The proportion of eyes gaining ≥15 letters at 148 weeks ranged from 35.8% to 42.9% for aflibercept versus 13.6% to 18.9% for laser control.

In the VIVID-East study of Asian patients with DME, both aflibercept regimens (2q4 and 2q8) achieved mean BCVA gains of +13.6 and +13.1 letters respectively at 52 weeks, compared to -0.5 letters with laser (p<0.0001). A significantly higher proportion achieved ≥15-letter gains (43.3% for 2q4, 36.5% for 2q8, versus 12.1% for laser).

When compared to other anti-VEGF agents, aflibercept showed similar effectiveness. In the comparison of aflibercept monotherapy versus bevacizumab-first with potential switch to aflibercept, mean improvement was 15.0 letters for aflibercept versus 14.0 letters for bevacizumab-first (adjusted difference 0.8 letters, 95% CI -0.9 to 2.5, p=0.37). The real-world retrospective analysis found that at 3 years, the adjusted mean visual acuity change was +2.4 letters for aflibercept versus +1.3 letters for ranibizumab (p=0.001).

Anatomical Outcomes

Aflibercept produced significant anatomical improvements in both wet AMD and DME. In wet AMD, the proportion of eyes achieving dry retina (absence of fluid on OCT) was similar between aflibercept and ranibizumab (RR 1.06, 95% CI 0.98 to 1.14). Choroidal neovascularization area reduction showed no significant difference between the two agents (mean difference -0.24 mm², 95% CI -0.78 to 0.29). In one observational study, the proportion of active CNV lesions decreased from 80% to 58% over 12 months after switching to aflibercept.

For DME, central retinal thickness reductions were substantial and significantly greater than laser photocoagulation:

The real-world comparison to ranibizumab showed aflibercept had significantly better anatomical outcomes, with adjusted mean central subfield thickness change remaining significantly different throughout the 3-year period (-114.4 μm for aflibercept versus -87.8 μm for ranibizumab, p<0.01).

Diabetic retinopathy severity also improved with aflibercept treatment. Greater proportions of eyes treated with aflibercept showed improvement of ≥2 steps in the Diabetic Retinopathy Severity Scale compared to laser control in both VISTA (29.9% for 2q4, 34.4% for 2q8, versus 20.1% for laser, p=0.0350 and p=0.0052 respectively) and VIVID (44.3% for 2q4, 47.8% for 2q8, versus 17.4% for laser, p<0.0001 for both). The proportion of patients with improvement >2 levels in diabetic retinopathy severity was significant for aflibercept groups versus laser.

Vision-Related Quality of Life

Quality of life assessments demonstrated clinically meaningful improvements with aflibercept treatment. In wet AMD, the NEI VFQ-25 scores showed similar improvements across all subscales over 52 weeks for aflibercept 2q8 and ranibizumab, with the greatest improvements in mental health and general vision (9.0-11.6 points for both treatments). Improvements of ≥4 points were observed for near vision, distance vision, role difficulties, and dependency subscales. Mean change from baseline in composite NEI VFQ-25 score showed meaningful improvement only in patients who gained ≥5 ETDRS letters (7.3 points for aflibercept 2q8 and 7.8 points for ranibizumab).

For DME, the AQUA study reported substantial quality of life improvements at 52 weeks:

The improvements were more pronounced for near activities than for distant activities, suggesting significant impact on daily functioning. These quality of life improvements support aflibercept use in DME as they were clinically meaningful.

Safety Profile

The safety profile of aflibercept was generally favorable and comparable to other anti-VEGF agents. In wet AMD trials, ocular and systemic adverse events were similar across aflibercept and ranibizumab treatment groups. The Cochrane review reported that serious systemic adverse events were similar between aflibercept and ranibizumab (RR 0.99, 95% CI 0.79 to 1.25), while the risk of serious ocular adverse events was lower in the aflibercept group, though with an imprecise estimate (RR 0.62, 95% CI 0.36 to 1.07). However, the quality of evidence for adverse events was graded as moderate due to imprecision.

For DME, safety profiles were consistent across studies:

In the comparison of aflibercept monotherapy to bevacizumab-first, serious adverse events (52% versus 36%) and hospitalizations for adverse events (48% versus 32%) were more common in the aflibercept-monotherapy group. The real-world meta-analysis indicated no new safety signals were identified in long-term use, though comprehensive safety data were limited. The most common serious ocular adverse event in the AQUA study was endophthalmitis, occurring in 0.5% of patients.

Long-term safety over 148 weeks showed that the incidence of adverse events remained consistent with the known safety profile of aflibercept. Overall, 53.6% of DME patients experienced treatment-emergent adverse events, with 26.8% being ocular events in the study eye.

Treatment Regimens

Aflibercept was administered at a dose of 2 mg across all studies. Two primary dosing regimens emerged:

Every 4 weeks regimen (2q4): Monthly injections throughout the treatment period. Every 8 weeks regimen (2q8): The more common regimen involved 3-5 initial monthly loading doses followed by maintenance dosing every 8 weeks. Specifically, wet AMD studies used 3 initial monthly doses, while DME studies used 5 initial monthly doses.

The eight-week dosing regimen represented reduced treatment requirements compared to monthly dosing regimens and offered the potential to reduce treatment burden and risks from frequent injections. Over 2 years in real-world practice, patients received a mean of 12.34 injections, with a reduction in injections in Year 2 versus Year 1. In the AQUA study, patients received a mean of 8.8 injections over 52 weeks. One observational study showed a decrease in mean number of injections from 7.4 in the 12 months before switching to aflibercept to 6.6 in the 12 months after switching.

Rescue treatment protocols allowed flexibility in dosing. From week 24 in some DME trials, aflibercept patients could receive active laser if rescue criteria were met, while laser control patients could receive aflibercept 2q8. From week 100, laser control patients who had not received rescue treatment could receive aflibercept as needed per retreatment criteria. The treat-and-extend regimen in real-world practice achieved a mean BCVA gain of +5.91 letters at 2 years.

Comparative Effectiveness

Aflibercept demonstrated non-inferiority to ranibizumab in wet AMD and superiority to laser photocoagulation in DME, while showing comparable effectiveness to bevacizumab.

Versus Ranibizumab in Wet AMD:

In the VIEW trials, aflibercept regimens were noninferior and clinically equivalent to ranibizumab, with all aflibercept groups within 0.5 letters of ranibizumab for mean change in BCVA. Visual acuity outcomes were similar at one and two years, with approximately 32% and 31% of patients in both groups gaining ≥15 letters respectively. The proportion of eyes achieving dry retina was similar (RR 1.06, 95% CI 0.98 to 1.14), and CNV area reduction showed no significant difference (mean difference -0.24 mm², 95% CI -0.78 to 0.29). The similarity in morphological outcomes and visual acuity, combined with aflibercept’s less frequent dosing schedule, supported its use as an alternative to ranibizumab.

Versus Laser in DME:

Aflibercept demonstrated statistically significant superiority over laser photocoagulation in improving BCVA and reducing DME. Mean BCVA gains were significantly higher for aflibercept groups (10.5-12.5 letters) compared to laser (0.2-1.2 letters). Significant improvements were also seen in central retinal thickness reductions and the proportion of patients gaining ≥15 letters. At 148 weeks, aflibercept groups showed sustained superior outcomes in BCVA gain and DRSS score improvements compared to laser control. In the VIVID-East study of Asian patients, aflibercept showed significant improvements in BCVA and central retinal thickness reduction compared to laser.

Versus Bevacizumab in DME:

In the direct comparison of aflibercept monotherapy versus bevacizumab-first with potential switch to aflibercept, no significant difference in visual outcomes was found over 2 years (mean improvement 15.0 versus 14.0 letters, adjusted difference 0.8 letters, p=0.37). This indicated non-inferiority of aflibercept compared to a bevacizumab-first strategy.

Versus Ranibizumab in DME:

The real-world retrospective analysis found that visual acuity change was not clinically different between aflibercept and ranibizumab, but aflibercept had better anatomical outcomes. The adjusted mean central subfield thickness change remained significantly different throughout the 3-year period in favor of aflibercept (p<0.01). When baseline visual impairment was moderate (VA ≤68 ETDRS letters), aflibercept showed faster improvement in visual acuity up to 18 months compared to ranibizumab, though outcomes were similar by 36 months.

Real-World Versus Clinical Trial Results:

The real-world meta-analysis reported mean BCVA gains of +4.49 ETDRS letters at 2 years, which was higher than ranibizumab’s real-world results (1.2 to 3.4 ETDRS letters), though direct comparative effectiveness data between aflibercept and ranibizumab in real-world settings at 2 years were not available. The switching study showed that when patients transitioned from ranibizumab to aflibercept, mean visual acuity remained unchanged but with a modest increase in treatment intervals, suggesting real-world effectiveness similar to clinical trial results.