Mohammad AskarPublic working copy · medical review pending

Geographic atrophy · 231

Geographic atrophy in AMD

Geographic atrophy is late age-related macular degeneration with loss of RPE, photoreceptors and choriocapillaris; define it with the CAM cRORA criteria, not with a single RPE-loss size.

Public working copy · medical review pending. Source, translation and medical checks remain open. Read the limitations in the text and references. These educational materials do not replace individual medical assessment. Published: 2026-11-18T19:45:52+01:00. Internal project references lead to this notice. They are not publicly accessible primary sources; the source limitations remain stated in the text.

Use multimodal imaging for diagnosis and follow-up, counsel functional limits honestly, and judge complement inhibitors and photobiomodulation by endpoint, jurisdiction and safety rather than by headline percentages.

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.

01Define atrophy with the CAM nomenclature

Geographic atrophy is late age-related macular degeneration in which the retinal pigment epithelium, photoreceptors and choriocapillaris are lost. It is an imaging-based clinical entity, not a diagnosis made by one RPE-loss measurement. 2

CAM cRORA requires three OCT criteria together: choroidal hypertransmission of at least 250 µm in any lateral dimension, corresponding RPE attenuation or disruption of at least 250 µm, and overlying photoreceptor degeneration. An RPE tear must be excluded. 2

iRORA describes incomplete cRORA criteria in AMD with conventional drusen: hypertransmission with corresponding RPE attenuation or disruption and overlying photoreceptor degeneration, but not all cRORA criteria. It is not simply a smaller cRORA, and it is not any single missing line on OCT. 3

iORA and cORA describe outer retinal atrophy with relatively preserved RPE. Relatively preserved is not a guarantee that the RPE is healthy, and both are distinct from cRORA. 3

The CAM nomenclature is a consensus for description and reporting, not by itself evidence of treatment benefit or a prognosis for an individual eye. The 208 export's historical 175 µm threshold is not a universal GA criterion, and fundus autofluorescence, near-infrared and colour imaging corroborate extent without replacing the OCT criteria. 1 2

02Use multimodal imaging without fixed brightness rules

Fundus examination may show sharply demarcated depigmented areas with visible larger choroidal vessels. Near-infrared imaging can help delineate lesions, while hypo-autofluorescence and peripheral hyper-autofluorescence need correlation with OCT rather than a fixed brightness rule. 2

OCT shows RPE and ellipsoid-zone loss, choroidal hypertransmission and inner retinal subsidence, the original waterfall sign. Inner sagging or a reflective bare Bruch membrane alone does not establish iRORA or cRORA. 2 3

Fundus autofluorescence supports extent and follow-up, but low signal can also be blocked, for example by blood, and a bright border is not specific proof of metabolic failure. Read it together with OCT and colour imaging. 2

Distinguish type 1, 2 and 3 macular neovascularisation and non-exudative degenerative cavities or tubulation that can mimic active fluid. OCT angiography shows flow without dye and does not demonstrate leakage. Reticular pseudodrusen lie above the RPE and mark increased progression risk, not a certain future outcome. 1 2

03Counsel risk and progression without a rigid timetable

The original GA estimates of at least 8 million symptomatic people, about 0.5% at 65-79 years, up to 11% in older cohorts, 16% legally blind and two-thirds unable to drive lack a traced population and definition. Extrafoveal lesions may be asymptomatic, and legal-blindness and driving criteria vary by jurisdiction. 1

Age, smoking, genetic background, fellow-eye disease, large or calcified and refractile drusen, reticular pseudodrusen and choroidal features inform risk. The original ancestry, sun-exposure, body-mass and inactivity associations should not be treated as deterministic causes or as a hierarchy for an individual. 1

The quoted 1.6 mm² per year growth is not a linear forecast: lesion size, multifocality, extrafoveal location and autofluorescence patterns including diffuse-trickling influence observed growth and the measurement scale. 1

The original five precursor categories, namely large drusen, reticular pseudodrusen, crystalline deposits, pigment epithelial detachment and vitelliform lesions, are imaging contexts rather than a required sequence or proof that reticular pseudodrusen are always the most dangerous phenotype. The attributed group and exact predictive model remain unverified. 1

Progression markers such as reticular pseudodrusen, multifocality, lesion size, extrafoveal location and autofluorescence pattern are associated with faster growth, not deterministic. Automated OCT segmentation is not a regulator-qualified primary endpoint without direct source support. 3

04Address function, low vision and cataract separately

Geographic atrophy can impair reading, mobility, contrast sensitivity and psychosocial wellbeing even when foveal-sparing visual acuity is still relatively preserved. Preserved acuity does not exclude disability. 1

Cataract surgery may improve the lens component of vision but does not erase an atrophic scotoma. Separate lens opacity from retinal atrophy before promising benefit. 1

Low-vision rehabilitation should address reading, mobility, contrast function and psychosocial support, and referral should be considered early. Patients should self-monitor for new distortion and seek prompt assessment for possible neovascular conversion. 1

Functional endpoints differ from anatomical lesion area, and an extrafoveal lesion can be asymptomatic while still progressing. Discuss expectations in terms of tasks, not only visual acuity. 1 2

05Read complement inhibition by endpoint and jurisdiction

Pegcetacoplan (Syfovre, C3) and avacincaptad pegol (Izervay, C5) entered United States GA care in 2023, which does not mean availability or authorisation everywhere. The original 20-25% slowing summarises selected anatomical outcomes across different trials, arms and time points, not guaranteed visual improvement or a universal effect. 1

OAKS and DERBY were 24-month multicentre double-masked sham-controlled trials in participants aged at least 60 years: 1258 were randomised, 637 and 621, and the modified intention-to-treat populations were 614 and 597. 4

Pegcetacoplan 15 mg monthly or every other month was compared with pooled sham, with lesion area on autofluorescence as the primary endpoint at 12 months and visual function as a secondary endpoint at 24 months. Both trials were sponsored by Apellis. 4

At 12 months the OAKS reductions of 21% and 16% were significant, whereas the DERBY reductions of 12% and 11% were not, with p values of 0.062 and 0.085. A non-significant primary endpoint in one trial and a positive one in the other is a genuine inconsistency, not a presentation detail. 4

At 24 months OAKS showed 22% (-0.90 mm², 95% CI -1.30 to -0.50) and 18% (-0.74, -1.13 to -0.36), and DERBY showed 19% (-0.75, -1.15 to -0.34) and 16% (-0.63, -1.05 to -0.22). 4

No differences were seen in key visual endpoints at 24 months. 4

New exudative AMD occurred in OAKS in 11%, 8% and 2% and in DERBY in 13%, 6% and 4%, for monthly, every other month and sham respectively. 4

GATHER2 was a 24-month phase 3 double-masked sham-controlled trial with 448 randomised participants, 225 active and 223 sham; one untreated sham participant was excluded from analyses, aged at least 50 years with non-centre-point GA and visual acuity from 20/25 to 20/320. 5

Monthly 2 mg avacincaptad pegol compared with sham gave square-root area growth of 0.336 versus 0.392 mm per year, a difference of 0.056 (95% CI 0.016-0.096, p 0.0064), or 14% relative. 5

This is not mm² per year and is not comparable head-to-head with pegcetacoplan. The trial was funded by Iveric Bio, an Astellas company. 5

In GATHER2, macular neovascularisation occurred in 7% versus 4% and exudative disease in 5% versus 3%. Zero endophthalmitis or inflammation over 12 months is not proof of rare safety. 5

An updated 24-month report was located but not appraised, and its month-12 rerandomisation of 96 monthly and 93 every-other-month eyes is not the baseline arm size. 5

The European Medicines Agency re-examination confirmed a negative opinion for Syfovre, with a negative opinion on 19 September 2024 and refusal on 16 December 2024, because the anatomical effect had not established a favourable clinical benefit-risk balance. 1 6

This was not a cost-based rejection, and there is no EU approval. The statement that complement inhibitors have been available since 2023 therefore cannot be used as a German treatment standard; check the current jurisdiction-specific decision. 1 6

The EU application for avacincaptad pegol was filed under the name Izelvay and was withdrawn on 24 October 2024 for effectiveness and benefit-risk concerns. 7

This is separate from the Syfovre refusal and does not describe the United States product Izervay. United States authorisations from 2023 remain distinct from European availability, so current local prescribing must be checked before use. 7

06Put safety, injection burden and monitoring first

The selected 2025 United States Syfovre label warnings state that retinal vasculitis or retinal vascular occlusion may occur after the first dose and can cause severe vision loss, and that treatment should be discontinued if it occurs. This is sight-threatening; its incidence cannot be inferred from a warning alone. 8

The same label also warns of endophthalmitis, intraocular inflammation, retinal detachment, increased intraocular pressure and the risk of neovascular age-related macular degeneration. Monitoring after injection and clear instructions about warning symptoms are part of safe use. 8

Safety cannot be transferred from one agent to another. For Izervay only selected indexed pharmacokinetic text was available in this review, so no whole-label comparison and no class-wide identical safety profile are claimed; each current label has to be checked separately. 9

Monthly or every-other-month intravitreal injections with associated visits are a substantial burden, and modified intention-to-treat analyses with discontinuations limit how far an average effect applies to one patient. Complement inhibition increased exudative conversion in the trials, so new distortion, metamorphopsia, sudden visual loss or haemorrhage need urgent assessment. Follow-up is individualised, with no rigid progression timetable. 4 5

07Keep photobiomodulation in perspective

Photobiomodulation with Valeda is not established routine care for age-related macular degeneration. The July 2025 statement of the German Ophthalmological Society, the Professional Association of Ophthalmologists and the Retinological Society recommends use only in clinical trials because efficacy and safety remain uncertain, and it criticises the population, missing data and the clinical effect. 10

Nine sessions over three to five weeks describe a treatment protocol, not proof that mitochondrial activation improves vision. CE marking or device authorisation is not proof of clinical utility and not an indication for every form of geographic atrophy. 10

The same statement notes a United States De Novo authorisation on 4 November 2024 for a specified dry age-related macular degeneration population. That decision is regulatory clearance, not evidence that all extrafoveal geographic atrophy should be treated, and it does not establish photobiomodulation as standard care. 10

What are the main learning points of the geographic atrophy lecture?

Define GA with the CAM cRORA criteria, use multimodal imaging without fixed brightness rules, counsel functional limits honestly, and interpret complement inhibitors and photobiomodulation by endpoint, jurisdiction and safety.

Which findings require urgent reassessment?

New distortion, metamorphopsia, sudden visual loss or haemorrhage, because of possible neovascular conversion.

References and access status

  1. 1
    Immutable original export; 208 AMD, atrophy and AI lecture; historical attribution and recording unchecked

    Complete export read; original attribution and any recording or transcript not independently verified. Historical 208 appraisal reused; not new full-source access, and its historical 175 µm threshold is not a universal GA criterion.

  2. 2
    CAM report 3, Sadda et al. 2018; PMC11366072

    Selected definition section and Table 2 text read 18 November 2026; not the complete article, images, supplement or corrections.

  3. 3
    CAM report 4; PMC7218279

    Selected indexed iRORA definition paragraph only read 18 November 2026; not just any missing EZ or 250 µm size alone; iORA and cORA retain RPE.

  4. 4
    OAKS and DERBY pegcetacoplan trials; PMID 37865470

    Complete abstract read 18 November 2026; not the full report or supplement. Sponsored by Apellis; modified intention-to-treat and visual endpoints as reported in the abstract.

  5. 5
    GATHER2 avacincaptad pegol trial; PMID 37696275

    Complete abstract read 18 November 2026; not the full report. Funded by Iveric Bio, an Astellas company. Updated 24-month report located but not fully appraised; month-12 rerandomisation not baseline arm N.

  6. 6
    EMA Syfovre EPAR overview and refusal

    Current EPAR overview and Q&A read 18 November 2026; reassessment report not full. Negative opinion 19 September 2024, refusal 16 December 2024, overview last updated 30 October 2025.

  7. 7
    EMA Izelvay EPAR overview

    Overview read 18 November 2026; the EU application was filed as Izelvay and withdrawn 24 October 2024 for effectiveness and benefit-risk concerns; not identical to the Syfovre refusal and distinct from the United States name Izervay.

  8. 8
    FDA SYFOVRE label 2025

    Selected indexed warnings read 18 November 2026; not the whole label. Retinal vasculitis or occlusion may occur after the first dose with severe vision loss; discontinue if it occurs. Endophthalmitis, intraocular inflammation, retinal detachment, intraocular pressure and neovascular risk are listed.

  9. 9
    FDA IZERVAY label 2026

    Located and selected indexed pharmacokinetic text only read 18 November 2026; not a whole clinical prescribing appraisal, so no whole-label or class-wide safety comparison is made.

  10. 10
    DOG, BVA and RG statement on photobiomodulation in AMD, 4 July 2025

    All three extracted pages read 18 November 2026; recommends clinical trials only and criticises population, missing data and clinical effect. Device authorisation or CE marking is not proof of clinical utility.

  11. 11
    DOG AMD 2023 guideline catalogue

    Catalogue only located 18 November 2026; full current recommendations pending. Historical 184 review had limited access; dates must be retained.

  12. 12
    Review 231 source Issue brief; new GA draft request

    Issue body and comments captured 18 November 2026; this is a new focused review, not an existing clinical lecture. Two recall questions and three patient paragraphs are new draft additions.