PULSAR supports selected extended aflibercept 8 mg regimens in treatment-naive nAMD, not universal superiority across nAMD, DME and RVO. Protocol, last assigned interval, visits and regional label are different measures.
Terminology and draft status
Terminology review is not complete. Selected actual WHO UMD queries and their limits are recorded in terminology.md; they do not approve the whole translation.
01Understand the shared pathway without merging diseases
VEGF contributes to permeability and abnormal vessel growth in nAMD, DME and RVO-associated macular oedema. Diabetes and venous disease can produce hypoxia and barrier dysfunction; nAMD also involves degenerative and RPE/choroidal processes. Inflammation, microglial activation, pericyte loss, endothelial junction changes and other factors such as PlGF interact. They are not one identical causal sequence in every eye, and not every oedema implies neovascularisation. The VEGF family includes VEGF-A, VEGF-B, VEGF-C and PlGF; alternatively spliced VEGF-A isoforms include VEGF-A165. Their shared name does not make every target or isoform clinically interchangeable. 1
The 5 medicines in the table differ in molecular form and target; the original also discusses the historical VEGF-A165-selective aptamer pegaptanib. Aflibercept is a receptor-Fc fusion, faricimab a bispecific antibody, and not every anti-VEGF agent is an antibody protein. Broader binding or a higher molar dose does not by itself establish greater visual benefit; an efficacy ceiling is an interpretation, not a universal biological limit. 1
Aflibercept binding to galectin-1 was demonstrated in 2015 biochemical/cellular experiments and investigated in proliferative diabetic retinal tissue. This is distinct from proving an additional clinically important treatment effect in PULSAR nAMD or treating all molecular targets as separately validated indications. The galectin entry is retained with that translational limit. 1 7
| Medicine | Molecular form | Targets and clinical limit |
|---|---|---|
| Bevacizumab (Avastin) | Complete IgG antibody. | VEGF-A isoforms; ophthalmic use from this product is off-label in the relevant context. |
| Ranibizumab (Lucentis) | Fab antibody fragment. | VEGF-A isoforms. |
| Brolucizumab (Beovu) | Single-chain antibody fragment (scFv). | VEGF-A isoforms; consider inflammatory/occlusive vascular safety. |
| Aflibercept (Eylea) | Receptor-Fc fusion protein. | VEGF-A, VEGF-B, PlGF; galectin-1 binding is experimental translational evidence, not proven extra nAMD benefit. |
| Faricimab (Vabysmo) | Bispecific antibody. | VEGF-A and Angiopoietin-2 (Ang-2); indication-specific clinical evidence needed. |
02Measure burden and interpret long-term comparisons cautiously
Frequent injections burden patients, carers and services through travel, anxiety and appointments. In the original nAMD Barometer claim, 92% of German ophthalmologists considered frequency potentially excessive, while 74% of patients valued longer intervals without visual loss: these are different questions, not the same response in two populations. The exact questionnaire, sample and date remain unresolved; assess each patient’s actual barriers. 1
The original workforce claims refer to Germany in 2020: 59% of ophthalmologists older than 50 and growth in practices of 0.4%, with the specialty described as the oldest. The precise workforce dataset and definition of growth remain unverified; these are historical claims, not current staffing facts. Population ageing and service capacity matter, but these figures do not directly quantify future injection needs. 1
A two-region 10-year registry comparison found mean change−0.9 letters in 132 completing ANZ eyes versus−14.9 in 37 Swiss eyes, with median 53 versus 42 injections. Regimen was inferred as more T&E versus PRN from visit/injection patterns; treatment was not randomised. Attrition, baseline differences and regional practice confound causal interpretation. The original “stable versus 15-letter loss” and almost-double injections in some years describe this context, not proof that one schedule guarantees an individual outcome. 1 8
03Separate a concentration model from a clinical interval
In a simple first-order decay model, doubling the starting concentration adds one half-life above an unchanged activity threshold and quadrupling adds two. The lecture’s7–11 day half-life, approximately 10 day extra for doubling and 20–22 days for 8 mg versus 2 mg are illustrative assumptions, not universal ocular kinetics for every anti-VEGF drug or patient. Tissue binding, clearance, dose volume and biological activity complicate the model. 1
Aflibercept 8 mg uses 0.07 mL in the studied formulation; greater duration must be demonstrated clinically rather than calculated from dose alone. Distinguish the interval between active injections from monitoring visits and loading doses. A12-,16-,20- or 24-week assignment in a trial is neither a promise of that interval for every eye nor automatic regional prescribing permission. 1 5
04Read PULSAR as a non-inferiority trial in naive nAMD
PULSAR randomised 1011 treatment-naive adults with nAMD;1009 received treatment. After 3 monthly doses in all arms,8 mg every 12 or 16 weeks was compared with 2 mg every 8 weeks. Disease-driven shortening could begin at week 16, with protocol-guided extension later. Randomisation and double masking strengthen inference, while the fixed 2q8 comparator does not compare all flexible real-world strategies. Bayer/Regeneron funded the trial. 1 2 3
At 48 weeks, mean gains were 6.7/6.2/7.6 letters for 8q12 / 8q16 / 2q8. Adjusted differences versus 2q8 were−0.97 letters (95% CI−2.87 to 0.92) and−1.14 (−2.97 to 0.69), meeting the 4-letter non-inferiority margin. This establishes the prespecified comparison, not superior vision, exact equality or equal rare-event safety. 2
At 96 weeks 869 participants completed treatment. Model-based LS mean gains were 5.6/5.5/6.6 letters, while completer mean injection counts were 9.7/8.2/12.8. The original 3.1 and 4.6 fewer injections are the respective differences from 12.8, not injection savings proven for every patient. Attrition and differences between modelled efficacy and completer denominators matter. 1 3
Among completers,87% of 8q12 had a last assigned interval≥12 weeks; in 8q16, 78%/53%/31% qualified for≥16/≥20/24 weeks. Last assignment differs from maintaining an interval throughout follow-up. The original 75%/70% maintained-without-shortening figures have unresolved definitions and denominators and should not be interchanged with these endpoint percentages. 1 3
05Keep anatomical response and safety in perspective
All arms reduced retinal thickness. The original week 16 drying result of 63% versus 52% refers to pooled 8 mg versus 2 mg central IRF+SRF freedom after loading; selected sponsor text previously accessed reports p=0.0002 from a one-sided analysis. The original week 48 figures of 69% versus 59% are retained but their exact endpoint, denominator and table are not yet verified. Dryness does not mean absence of every retinal abnormality or superior vision. 1 4
The post-hoc 80% versus 66% example associates dryness at week 4 with maintaining≥16-week dosing through 48 weeks. Its analysis population and definitions remain unverified; association is not a validated individual decision rule. Baseline acuity or thickness also cannot be declared completely non-predictive from a lack of a demonstrated subgroup interaction. Use repeated clinical activity, not a single early image, to guide extension. 1
Similar trial safety and no identified new signal do not prove the absence of rare inflammation, occlusive vasculitis, endophthalmitis or pressure complications. Larger 0.07 mL volume and IOP still require monitoring. Drug-specific risks also differ: brolucizumab has a recognised retinal vasculitis/vascular-occlusion concern, so longer action is not a reason to treat every agent as equally safe. 1 5
06Plan monitoring and discuss cessation using the correct denominator
In a registry study of 434 eyes with treatment suspended after at least 3 months of documented inactivity, estimated reactivation was 41% at 1 year and 79% at 5 years. Median reactivation time was 504 days, not the original 192-day figure, whose source remains unresolved. Among 206 reactivated eyes with at least 12 months of resumed treatment, mean net loss was 3.3 letters (95% CI 2.3–5.1). This selected observational study does not give every patient the same risk or establish what 8 mg cessation would do. 1 6
Plan proactive disease-guided care with the patient, checking function, OCT fluid, haemorrhage and adverse effects. Treatment cessation, if considered because of sustained inactivity, end-stage damage, burden or changing goals, needs an individual discussion and monitoring plan; neither unmonitored stopping nor indefinite injections regardless of benefit follows from these studies. Disease reactivation may not recover fully after retreatment. 1 6
07Keep trial findings, indication and local prescribing separate
PULSAR concerns nAMD; DME needs PHOTON and RVO separate evidence. Current accessed EU Eylea 114.3 mg/mL information lists adult nAMD, DME and RVO-related oedema, with 8 mg in 0.07 mL; the schedules are indication-specific. For initiating nAMD/DME treatment, 3 monthly doses precede response-guided extension up to 4 months, then potentially up to 6 months with stable visual/anatomical outcomes. These are permitted selected schedules, not guaranteed intervals for every eye. 1 5
For a switch to EU Eylea 8 mg, stable visual/anatomical outcomes may allow maintenance or extension after the first dose; suboptimal response may warrant up to 3 monthly doses before adjustment. Thus “switch without reloading” is not universal. Assess function, fluid, haemorrhage, safety, current product information and preferences, distinguish injection intervals from visits, and document the next review. Longer intervals may reduce burden but do not themselves prove improved adherence or quality of life. 1
What are the main learning points?
PULSAR supports selected extended aflibercept 8 mg regimens in treatment-naive nAMD, not universal superiority across nAMD, DME and RVO. Protocol, last assigned interval, visits and regional label are different measures.
When is assessment urgent?
New distortion, visual loss or haemorrhage needs prompt assessment. Severe pain, redness, light sensitivity or marked visual deterioration after an injection needs immediate eye-emergency assessment. Do not change or stop injections on this page’s interval examples. Rapid visual loss, severe pain, marked inflammation, a pressure rise or suspected infection or neurological disease also needs urgent assessment.
References and access status
- 1Immutable original export; longer-acting retinal treatment synthesis
Complete export read; original attribution and any recording/transcript or appointment not independently verified.
- 2PULSAR 48 week phase 3 trial. PMID 38461841
Complete primary NCBI abstract independently reread 15 November 2026; full article/protocol/figures/supplements unread. Gillies abstract has an unresolved 712 versus 474+321 arithmetic discrepancy.
- 3PULSAR 96 week trial report. PMID 40876598
Complete primary NCBI abstract independently reread 15 November 2026; full article/protocol/figures/supplements unread. Gillies abstract has an unresolved 712 versus 474+321 arithmetic discrepancy.
- 4Sponsor PULSAR 48 week drying poster
Reused 10 November indexed text/captions only, not figure pixels or full post-hoc analysis.
- 5EMA Eylea EU product information: selected 8 mg sections
Actual downloaded current official PDF selected 8 mg sections 4.1–4.4 read 15 November 2026; pages 64–67. Not a full-label, trial-table or German reimbursement appraisal.
- 6Nguyen 2019: outcomes after suspending VEGF inhibitors. PMID 31281103
Complete primary NCBI abstract independently reread 15 November 2026; full article/protocol/figures/supplements unread. Gillies abstract has an unresolved 712 versus 474+321 arithmetic discrepancy.
- 7Aflibercept traps Galectin-1.2015, PMID 26648523
Complete primary NCBI abstract independently reread 15 November 2026; full article/protocol/figures/supplements unread. Gillies abstract has an unresolved 712 versus 474+321 arithmetic discrepancy.
- 8Gillies 2020:10-year nAMD outcomes from two regions. PMID 31606444
Complete primary NCBI abstract independently reread 15 November 2026; full article/protocol/figures/supplements unread. Gillies abstract has an unresolved 712 versus 474+321 arithmetic discrepancy.
For patients
Longer intervals can make treatment easier, but your eye’s response and follow-up determine whether extension is appropriate. Some eyes need shorter intervals again. Discuss travel, carers and appointment difficulties with the team.
Examination and selected tests guide individual treatment and follow-up; this educational text cannot replace your own assessment.
New distortion, visual loss or haemorrhage needs prompt assessment. Severe pain, redness, light sensitivity or marked visual deterioration after an injection needs immediate eye-emergency assessment. Do not change or stop injections on this page’s interval examples. Rapid visual loss, severe pain, marked inflammation, a pressure rise or suspected infection or neurological disease also needs urgent assessment.
New distortion, visual loss or haemorrhage needs prompt assessment. Severe pain, redness, light sensitivity or marked visual deterioration after an injection needs immediate eye-emergency assessment. Do not change or stop injections on this page’s interval examples. Rapid visual loss, severe pain, marked inflammation, a pressure rise or suspected infection or neurological disease also needs urgent assessment.