PULSAR supports non-inferior visual outcomes with fewer injections at 48 weeks under its protocol, not a promise of a particular interval or superiority. Interval figures, drying endpoints and switch claims need their own denominators and limits.
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.
01Treatment burden and demographic pressure
Anti-vascular endothelial growth factor (anti-VEGF) treatment has changed neovascular age-related macular degeneration (nAMD) over the past 20 years: in most patients it can preserve, and often improve, visual acuity. The practical question has therefore shifted from whether to treat to how to deliver effective treatment with less burden for patients, carers and clinics. 1
The original briefing reports a shared wish for fewer visits. In a global Barometer survey, 73% of surveyed patients said longer intervals without vision loss mattered to them, 88% of ophthalmologists thought treatment frequency could be too high, and nearly half of patients found the burden too great. The survey source, dates, selection and denominators are not verified here; these are reported experiences, not measured population rates. 1
The practical barriers are concrete. In rural regions such as the Mecklenburgische Seenplatte, a one-way journey can take up to 1.5 hours; transport costs may not be reimbursed, companions must be organised, and parking adds cost. Patients also worry about burdening family and friends and may fear the injection itself. These travel, carer, cost and anxiety examples describe regional experience, not a measured national dataset. 1
Clinicians face the same pressure from the other side: frequent treatment consumes appointment slots, staff time and clinic capacity, and 88% of ophthalmologists in the original survey considered frequency potentially excessive. 1
Demographic change adds a second strain. The ageing population is projected to raise AMD prevalence in Europe from about 67 million in 2015 to 77 million by 2050; this figure concerns all AMD, not nAMD alone, and the 2015 baseline attribution is not confirmed here. At the same time, nearly 60% of German ophthalmologists are said to be at least 50 years old and due to retire within 10-15 years; that workforce figure also lacks a verified source and denominator. The original concludes that innovative solutions and care models are needed for adequate care with reduced effort, so this is a care-model problem, not only a drug problem. 1 5
02Variable response: a teaching model, not a measured distribution
Response to anti-VEGF treatment varies widely. The original describes a bipolar pattern with two peaks rather than a normal distribution: chronically active CNV (CAC), in which residual fluid persists despite regular treatment and injections are needed often, and effectively controlled CNV (ECC), in which a few treatments produce a durably dry retina that can be followed at long intervals. This is a useful teaching heuristic for discussing heterogeneity, but it is not a validated population classification, and the supplied material does not demonstrate the claimed curve. 1
The stated aim of modern therapy is to shift this distribution to the right, moving more patients into the long-interval group. That is a conceptual goal. It is not a validated outcome measure, and none of the supplied data show that a drug changes an underlying two-peak population distribution. Judge an individual eye by its activity, not by an assumed class. 1
For practice, the model is a reminder that trial means conceal wide variation. Two eyes with the same baseline acuity can behave very differently, and a mean gain of a few letters says little about the next patient in the clinic. Acuity, OCT fluid, haemorrhage, lesion activity, adherence and the reliability of follow-up all belong in the interval decision. 1 3
03PULSAR: what was actually randomized
PULSAR was a global, randomized, double-masked phase III trial in treatment-naive adults aged 50 years or older with nAMD. It randomized 1011 people and treated 1009. After three initial monthly injections, aflibercept 8 mg every 12 weeks (8q12) or every 16 weeks (8q16) was compared with aflibercept 2 mg every 8 weeks (2q8); disease-based shortening was possible from week 16, and follow-up ran to 96 weeks. The original phrase 'more than 1,000 patients' refers to this trial population. 2
At 48 weeks, mean best-corrected visual acuity (BCVA) improved by +6.7 letters with 8q12 and +6.2 with 8q16, versus +7.6 with 2q8. The adjusted differences versus 2q8 were −0.97 letters (95% CI −2.87 to 0.92) and −1.14 (−2.97 to 0.69), respectively. Both met the prespecified 4-letter non-inferiority margin; this does not establish superiority or identical regimens. 2
Randomization and double masking strengthen the comparison; Bayer and Regeneron funded the trial. Missing-data assumptions, selective reporting, multiplicity, subgroup interactions and rare harms still need full-report appraisal. The trial population also limits direct application to previously treated eyes. 2 3
At 96 weeks, 869 of 1009 treated participants completed treatment. The least-squares mean BCVA changes were +5.6 letters (95% CI 4.1–7.1), +5.5 (4.0–7.0) and +6.6 (5.2–8.0) for 8q12, 8q16 and 2q8. These are model-based efficacy estimates; the US label reports them for the full analysis set. Separately, completers received mean injection counts of 9.7, 8.2 and 12.8. Attrition and modelling assumptions matter; none is a guaranteed practice average. 3 6
04Results, drying endpoints and the limits of the table
The original presents the interval results as impressive potential for extension, but the numbers need their denominators. The 96-week abstract reports arm-specific last assigned intervals among completers. In 8q12, 87% were at an interval of at least 12 weeks. In 8q16, 78% were at least 16 weeks, 53% at least 20 weeks and 31% at 24 weeks. A last assigned interval is not the same as an interval maintained throughout the whole study, and a completer proportion is not a rate for all randomized patients. 3
The original instead gives pooled figures: 88% at an interval of at least 12 weeks, nearly half at least 20 weeks, and 75% in 8q12 and 70% in 8q16 never shortened. It also says intervals stayed stable in the great majority over the study course despite the abrupt extension after loading. These pooled and never-shortened values cannot be reconstructed from the accessed abstract; the pooling, definitions and denominators are unresolved. They are retained as unverified original claims, not converted into new verified population rates. 1 3
The drying result needs precise wording. At week 16, after three monthly doses, the sponsor-indexed key secondary endpoint was absence of both intraretinal and subretinal fluid in the central subfield: 63% with pooled 8 mg versus 52% with 2q8, p=0.0002, tested one-sided. This is not whole-retina dryness, and it does not prove proportionately greater visual benefit. 1 7
The original also links dryness at week 4 to maintaining a 16-week interval through week 48: about 80% versus 66% in patients who were never fluid-free in the early phase. That analysis was exploratory, and the association is not verified as a validated selection rule or an individual prediction. Importantly, about two-thirds of patients with early residual fluid still reached the 16-week interval, so early fluid must not be treated as automatic failure. 1 7
Table 1 keeps the three original rows and adds the trial bounds. It separates the 48-week non-inferiority result, the injection counts and the 96-week visual stability, without turning any row into a promise about an individual patient. 2 3
| Metric | Result |
|---|---|
| Primary endpoint (week 48) | Non-inferiority met: mean BCVA gain +6.7 letters with 8 mg q12 and +6.2 with q16 versus +7.6 with 2 mg q8; non-inferiority margin 4 letters. No superiority or equality claim. |
| Number of injections | Clearly fewer injections in the 8-mg arms; among completers through week 96, mean 9.7 (q12) and 8.2 (q16) versus 12.8 (q8). |
| Visual stability (week 96) | Model-based BCVA gains +5.6/+5.5/+6.6 letters (8q12/8q16/2q8, full analysis set). Separately, 869 of 1009 treated participants completed treatment; injection counts describe completers. |
05Extension, switching and subgroup claims
The open-label extension followed patients to week 156. Visual gains and reduced retinal thickness were reported to be maintained over three years with a lower treatment burden. Patients originally in the 2-mg arm were switched to 8 mg after week 96 without a new loading course; they reportedly kept stable vision and long intervals, and 8 in 10 reached at least 12 weeks. The extension was open-label and selected, the complete figures and denominators were not appraised, and there was no unchanged randomized comparator for the switch. 1 8
These data do not directly answer the common clinical question about a patient who is already receiving treatment elsewhere. PULSAR enrolled treatment-naive eyes, so its randomized comparison is not evidence that switching a previously treated eye produces the same benefit. The original briefing goes further and states that both treatment-naive and previously treated patients benefit equally; that claim is not supported by the randomized trial. The two teaching cases later in this lecture illustrate feasibility in individual patients; they are not comparative trials. 1 2
The original states that all examined subgroups benefited equally, regardless of baseline BCVA, retinal thickness, CNV type or size and ethnicity. Consistent direction across subgroups is reassuring, but without interaction tests, precision estimates and multiplicity control it does not prove identical effects, absence of heterogeneity or ethnic equality. A subgroup that looks similar can still differ meaningfully. 2 3
Prescribing information differs by jurisdiction and date. The US label revised in April 2026 allows 8–16-week intervals after three monthly doses, selected extension to 20 weeks after one year of successful response, and 4-week dosing for some patients. Earlier selected EU wording allowed intervals up to 6 months in stable eyes. That historical wording does not establish the current German indication: check the current local product information before prescribing. 6 9
06Safety, monitoring and the two teaching cases
The original claims that no endophthalmitis or occlusive retinal vasculitis occurred in the 8-mg arms over two years and that intraocular pressure and inflammation rates were comparable to 2 mg. That is an unverified study-table claim. Zero events in a trial never mean zero real-world risk. The current US label reviewed for this draft warns of endophthalmitis, retinal detachment, rare retinal vasculitis with or without occlusion, intraocular pressure increase and arterial thromboembolic events. Post-marketing rates for aflibercept in general must not be transferred to 8 mg as if they were product-specific. 1 6
Monitoring is therefore not optional. Before extending an interval, review acuity, OCT activity including intraretinal and subretinal fluid, haemorrhage, lesion stability and the reliability of follow-up. The original principle of extending only after complete disease control, with no intraretinal fluid, is a reasonable contextual goal; it is not a diagnostic trial proving that every small fluid pocket requires treatment. The statement that switching should be considered whenever intervals are no more than 8 weeks is an expert view, not a universal rule. 1 9
Case 1, the treatment-naive patient: a 73-year-old woman presented with a fresh CNV, acuity 0.2 and marked metamorphopsia. After a loading course of three 8-mg injections the retina was completely dry, acuity improved to 0.4 and the distortion disappeared. At the 8-week review, slight lesion activity had returned. The lesson is close, individual monitoring with controlled treat-and-extend rather than an immediate jump to the longest interval. This is one uncontrolled course, not a comparative effect. 1
Case 2, the previously treated patient: an 85-year-old man had persistent disease activity and acuity drops under off-label therapy at an 8-week interval. After switching to 8 mg and a new loading course, the lesion dried completely and the interval was stably extended to 12 weeks with acuity 0.8. The lesson offered is that high-dose treatment can be an option when standard treatment does not achieve sufficient control. Because the course is uncontrolled and prior treatment and selection are uncontrolled, it does not establish a general reloading or switching recommendation. 1
The original presents both cases as practical experience alongside the trial data. Taken together, they support a familiar clinical sequence: treat, confirm control, then extend gradually with OCT and acuity, and shorten again if activity returns. They do not support treating every patient identically, and they do not replace the trial evidence or the product information. 1 9
07Systemic perspectives: fire model, AREDS and saccharin
The lecture's final section treats AMD as more than an isolated eye disease. Genetic factors, nutrition with the AREDS connection, smoking, physical activity and immune factors may modulate the course. The image of systemic pro-angiogenic factors as oil on the fire of macular neovascularisation, while anti-angiogenic factors extinguish it, is a conceptual metaphor, not a validated model. AREDS supplementation is not a blanket rule for all AMD and is not a substitute for anti-VEGF treatment. Smoking remains an important environmental factor; exercise has experimental CNV associations, not proven treatment equivalence. 1
The saccharin study combined a cross-sectional analysis of 46 patients with nAMD and laser-induced CNV in mice. Measured saccharin metabolites were associated with better lesion control (p=.004) and absence of subretinal hyperreflective material (p=.015). This was not a randomized human dietary trial. Confounding, reverse causation, multiple testing and animal-to-human translation all limit the clinical inference,. 4
In the mouse model, less leakage, bleeding and fibrosis were reported. A 40% reduction in mononuclear phagocytes had p=.06, which is not statistically significant, and the TAS1R3 receptor proposal is mechanistic. These findings do not justify dietary advice. What would change practice is randomized human outcome data, not an association or a mouse experiment. 4
For the reader, the safe conclusion is to keep systemic risk factors in the overall conversation, avoid promising that a lifestyle change will replace injections, and tell patients honestly that the saccharin signal is preliminary. The lecture's systemic framing is a research direction; it is not a treatment recommendation. 1 4
What are the main learning points of this nAMD briefing?
PULSAR shows non-inferior visual outcomes at 48 weeks with fewer injections under its protocol, not a promise of a particular interval or superiority. Drying endpoints, pooled interval figures and switch claims each have separate limits; monitoring remains individual.
Which findings require urgent assessment?
Sudden visual deterioration, severe pain, marked inflammation, a rise in intraocular pressure or suspicion of an infectious or neurological cause require immediate medical assessment.
References and access status
- 1Original export: Briefing-Dokument: Patientenversorgung bei nAMD im klinischen Alltag erleichtern
Complete original export supplied by the parent reviewer and read for this draft. Survey sources, pooled interval figures, speaker attribution and the sponsored extension report were not independently verified in this drafting pass.
- 2Intravitreal aflibercept 8 mg in neovascular age-related macular degeneration (PULSAR): 48-week results from a randomised, double-masked, non-inferiority, phase 3 trial.
Parent-reported access on 10 November 2026: complete actual NCBI abstract read; full report, protocol and supplement unread. Not independently re-opened in this drafting pass.
- 3Intravitreal Aflibercept 8 mg in Neovascular Age-Related Macular Degeneration: Ninety-Six-Week Results from the Randomized Phase 3 PULSAR Trial.
Parent-reported access on 10 November 2026: complete actual NCBI abstract read; completer and arm-specific intervals distinguished. Full report and 156-week extension unread. Not independently re-opened in this drafting pass.
- 4Exploring the Impact of Saccharin on Neovascular Age-Related Macular Degeneration: A Comprehensive Study in Patients and Mice.
Parent-reported access on 10 November 2026: complete actual NCBI abstract read; cross-sectional plus animal model, not a human dietary RCT. Full report unread. Not independently re-opened in this drafting pass.
- 5Prevalence and incidence of age-related macular degeneration in Europe: a systematic review and meta-analysis.
Parent-reported access on 10 November 2026: complete actual NCBI abstract read; the 67 to 77 million projection concerns all AMD, not nAMD. Full report unread. Not independently re-opened in this drafting pass.
- 6Eylea HD (aflibercept) US prescribing information
Parent-reported access on 10 November 2026: selected sections 2.2 and 5.1-5.3 and 14.1 extracted text read; not the complete label and not a German or EU prescribing authority.
- 7Sponsor poster: PULSAR 48-week drying results (SOG 2023)
Parent-reported access on 10 November 2026: indexed sponsor poster result and caption extract only; exact figure images were not inspected.
- 8Sponsor encore presentation: PULSAR week 156 (ASRS 2025)
Parent-reported access on 10 November 2026: title, disclosures and indexed week-156 summary only; graphics, denominators and complete report unread. A local PDF attempt returned HTML and the screenshot did not expose images; no visual appraisal.
- 9EMA Eylea EU product information
Selected printed pages 64-68 and sections 4.1-4.4 for 8 mg were actually inspected in earlier work on 7 November 2026. The 10 November 2026 attempt returned HTTP 429/HTML, so this is not fresh current-label verification; not the complete label.
For patients
What does this mean? Wet AMD can impair central vision. Injections and follow-up aim to control disease activity, and the appropriate interval is chosen individually.
Examination and treatment: this information supports orientation and does not replace an individual examination by an ophthalmologist. Do not change or omit treatment on your own because of a study or a dietary hypothesis.
When should you act quickly? Seek immediate medical help for sudden visual deterioration, severe pain or a new visual field defect.
Seek immediate medical help for sudden visual deterioration, severe pain, marked inflammation, a rise in intraocular pressure or a new visual field defect.