Age-related macular degeneration (AMD) is a disease that causes atrophic or exudative changes in the macula due to age-related changes. It is considered a multifactorial disease involving genetic predisposition and environmental factors such as aging, smoking, sunlight exposure, obesity, and high-fat diet.
It is defined as age-related macular abnormalities within a 6,000 μm diameter centered on the fovea in individuals aged 50 years or older, and is broadly classified into precursor lesions, neovascular type, and atrophic type. It is the fourth leading cause of visual impairment.
According to the Hisayama Study (2012), the prevalence of advanced AMD is 1.6% (exudative type 1.5%, atrophic type 0.1%), and the Nagahama Study (2008–) reported a prevalence of precursor lesions of 22.8% (drusen 39.4%). It occurs in men and women aged 50 years and older (male:female = 3:1), affecting one or both eyes (approximately 40%). Both precursor lesions and advanced AMD are increasing. The Hisayama Study (9-year follow-up) reported that smoking increased the incidence of late AMD by 4-fold 1).
There are racial differences, with higher prevalence in Caucasians and Asians, and lower in Hispanics and Africans 2). The prevalence of atrophic AMD is reported to be 0.66–1.34% in Western studies, and in those aged 85 years and older, it is said to be 4 times more frequent than exudative AMD. This is likely to become a major problem in Japan as the population ages. Worldwide, approximately 200 million people have AMD, and it is predicted to increase to about 288 million by 2040 2). The prevalence of late AMD increases exponentially with age, from 0.1% in those aged 50–59 years to 4.3% in those aged 80 years and older 2).
The latest Japanese clinical practice guidelines (2024) classify AMD into the following four stages based on the Beckman classification 1).
| Stage | Characteristics |
|---|---|
| Early AMD | One or more medium soft drusen (63 to <125 μm) |
| Intermediate AMD | Large drusen (≥125 μm), RPE abnormalities, subretinal drusenoid deposits |
| Late AMD | Macular neovascularization (MNV) (including PNV) or geographic atrophy |
| End-stage AMD | Fibrotic scar or cystoid macular degeneration with severe vision loss |
Small (hard) drusen (<63 μm) are considered physiological age-related changes and are not included in early AMD. However, if numerous (20 or more) hard drusen are present, the risk of developing AMD is high 1). The 5-year progression rate of intermediate AMD is approximately 18%, but the risk increases significantly in the presence of reticular pseudodrusen (pigment abnormalities + large drusen + reticular pseudodrusen: 5-year risk 72%) 2).
Neovascular AMD
Macular neovascularization (MNV): New blood vessels originating from the choroid or retinal vessels develop in the macula, causing sudden vision loss.
Types: Classified into type 1 MNV (sub-RPE), type 2 MNV (supra-RPE), type 3 MNV (retinal angiomatous proliferation, RAP), and PCV (type 1 MNV + polypoidal lesions) 1).
Characteristics in Japan: About half of neovascular AMD cases are pachychoroid neovasculopathy (PNV), and drusen are observed in only about 30% 1).
Atrophic AMD
Geographic atrophy (GA): Characterized by well-demarcated atrophy of the RPE, photoreceptors, and choriocapillaris.
Natural course: Begins around the fovea and expands in a horseshoe → ring shape. Growth rate is 1.28–2.6 mm²/year 3).
CNV development during course: 10–15% of cases develop choroidal neovascularization, transitioning to the exudative type.
Approximately 40% of cases develop in both eyes. If one eye has late AMD, the fellow eye has a high rate of MNV. The ARMS2 genotype has been reported as a predictor of fellow eye involvement 1). Regular eye examinations and self-checking with an Amsler grid are recommended.
Initially, it begins with metamorphopsia (distorted vision) and central scotoma. As it progresses, visual acuity drops to less than 0.1. In cases of massive hemorrhage, sudden severe vision loss may occur.
In cases of monocular onset, patients often do not notice symptoms in daily life. The severity of symptoms varies depending on the location and size of MNV, and the extent of subretinal fluid, hemorrhage, and fibrous scarring. In the neovascular type, the time to legal blindness tends to be shorter than in the atrophic type, and prompt consultation is recommended when metamorphopsia appears.
Yellow-white small round elevated lesions seen in the macula, consisting of polymorphous material (membranous debris, unesterified cholesterol, complement, etc.) accumulated between the RPE basement membrane and the inner collagenous layer of Bruch’s membrane. They are considered the origin of chronic inflammation, which is the core of AMD pathology.
Geographic atrophy is a well-demarcated area of RPE atrophy through which choroidal vessels are visible. Lesions begin around the fovea (juxtafoveal) and expand in a horseshoe → ring shape, progressing to the fovea. The median time to foveal involvement is 2.5 years3). Growth rate is 1.28–2.6 mm²/year; in ranibizumab trial data, GA area expanded from an average of 8.07 to 12.05 mm² over 2 years3). Multifocal GA has a faster growth rate than unifocal GA3). The conversion rate to GA in the fellow eye reaches approximately 30% at 12 months, and the conversion rate to CNV is reported as 6.7%.
Rapid atrophy may also occur when large drusenoid PEDs regress. Hyperautofluorescence patterns at the border of atrophy on fundus autofluorescence are useful for predicting progression rate.
Metamorphopsia and scotomas can be self-checked using an Amsler grid. However, early stages are often asymptomatic, making regular eye examinations essential. Especially in unilateral cases, it is not uncommon to be unaware in daily life.
AMD is a multifactorial disease, caused by a combination of genetic predisposition and environmental/behavioral factors.
AREDS2 supplements (lutein, zeaxanthin, vitamins C and E, zinc, copper) reduce the risk of progression from intermediate AMD to late AMD by about 25% 2). However, their effectiveness for early AMD or primary prevention has not been established. Beta-carotene increases the risk of lung cancer in smokers, so smokers should choose products containing lutein/zeaxanthin 1).
The diagnostic criteria for neovascular AMD are as follows 1):
The diagnostic criteria for atrophic AMD (geographic atrophy) are all of the following 1):
| Item | Criteria |
|---|---|
| Size | Diameter ≥250 μm |
| Shape | Round, oval, lobular, or geographic |
| Border | Sharp |
| RPE changes | Hypopigmentation or depigmentation changes |
| Choroid | Medium and large vessels clearly visible |
If exudative changes from MNV (IRF, SRF, sub-RPE fluid, fibrin, hemorrhage, etc.) are observed, it is judged as “active” 1). Non-invasive evaluation using OCT has become mainstream. Since the site of high disease activity in MNV is not necessarily in the fovea, scanning the entire macula for evaluation is recommended 1).
Macular abnormalities with MNV require differentiation from the following diseases.
OCTA has high accuracy for MNV detection with sensitivity 0.87 and specificity 0.97 2), and its use as a non-invasive test is increasing. It may be superior to fluorescein angiography especially for detecting type 1 MNV 1). However, ICGA remains essential for diagnosing PCV, and OCTA is not a complete replacement.
There is no evidence-based treatment for early AMD 2). For intermediate AMD or higher, smoking cessation, dietary improvement, and supplementation based on the AREDS2 formulation are recommended 1).
Supplement formulation used in AREDS2 1):
Beta-carotene was replaced with lutein/zeaxanthin because it increases the risk of lung cancer in smokers1). The AREDS2 formulation reduces the risk of progression from intermediate to late AMD by approximately 25%.
The first-line treatment for neovascular AMD is intravitreal injection of anti-VEGF agents1). For subfoveal MNV, anti-VEGF monotherapy is recommended as initial treatment.
Ranibizumab
Dose: 0.5 mg intravitreal injection
Loading phase: 3 injections at 1-month intervals
Maintenance phase: Pro re nata (PRN). In the MARINA study, visual acuity improved by 6.6 letters compared to a 14.9-letter decrease in the sham group1).
Biosimilar: Ranibizumab BS is available.
Aflibercept
Dose: 2.0 mg intravitreal injection
Loading phase: 3 injections at 1-month intervals
Maintenance phase: Fixed dosing every 2 months or treat-and-extend regimen. The interval is extended by 2 weeks (up to 3 months) if no exudative recurrence, and shortened by 2 weeks upon recurrence 1).
PCV: Polyp regression rate of 40–50%, superior to ranibizumab (20–30%).
Brolucizumab / Faricimab
Brolucizumab: 6 mg dose. Can be administered every 8–12 weeks. Caution required for intraocular inflammation including retinal vasculitis and vascular occlusion 1).
Faricimab: VEGF-A + Ang-2 bispecific antibody. Administered every 8–16 weeks. Non-inferior to aflibercept in the TENAYA/LUCERNE studies 1)10).
The following three dosing regimens are available 1).
For type 2 MNV or PCV, laser photocoagulation can be applied to the entire MNV. Use a wavelength of yellow or longer, spot size 200–300 μm, power 150–250 mW, duration 0.2–0.5 seconds, with a 100 μm safety margin around the MNV to achieve moderate or greater photocoagulation. However, laser photocoagulation irreversibly damages the RPE and is not suitable for MNV close to the fovea 1).
Treatment options for PCV are as follows 1).
PDT regimen: Intravenous infusion of verteporfin 6 mg/m² over 10 minutes. Laser irradiation (689 nm, 600 mW/cm², 83 seconds) is performed 15 minutes after the start of infusion. The spot size is the greatest linear dimension of the lesion plus 1,000 μm. Avoid direct sunlight for 2 days after treatment.
Long-term PDT may worsen macular atrophy; it should be avoided in cases with thin choroid or pre-existing macular atrophy. PDT is not recommended for type 3 MNV 1).
If the response to anti-VEGF therapy is poor (treatment resistance) or the effect diminishes (acquired resistance), switching to another agent may be effective 1). Drug changes may also be considered based on treatment burden (frequency of visits and injections). In end-stage AMD with inactive fibrotic scar or atrophic changes, active treatment is not indicated and observation is considered 1).
For eyes with good visual acuity, anti-VEGF monotherapy is an option; if reduction of treatment frequency is desired, combination therapy with PDT and VEGF inhibitors can be considered. In clinical practice, anti-VEGF monotherapy is the mainstream.
Massive submacular hemorrhage causes acute vision loss. If treated early, displacement of the hemorrhage may improve vision.
Currently, there is no established treatment for cases with well-established geographic atrophy involving the fovea. The complement system is thought to be strongly involved, and several molecular targeted drugs against the complement pathway are under development or in clinical trials.
For geographic atrophy outside the fovea, AREDS2 supplement intake and lifestyle improvements are recommended. If MNV develops during the course (10–15%), treatment with anti-VEGF drugs becomes standard.
For patients with progressive vision loss, low vision care such as recommending visual aids like sunglasses and magnifiers, and support for daily living is important 1).
In the induction phase, three injections are usually given at one-month intervals. In the maintenance phase, the treat-and-extend method (gradually extending intervals) is recommended. The ALTAIR study (Japanese subjects) confirmed efficacy over 96 weeks 1). With faricimab, some cases can maintain up to 16-week intervals 10).
Currently, there is no established treatment. In the US, two complement inhibitors were FDA-approved in 2023 9), but they are not yet approved in Japan. AREDS2 supplements reduce the risk of progression to late AMD but have not been shown to suppress the progression of geographic atrophy itself. When vision loss progresses, low vision care becomes important.
The pathology of AMD begins with damage to RPE cells. Drusen accumulate between the RPE basement membrane and the inner collagen layer of Bruch’s membrane. Drusen components include membranous debris, unesterified cholesterol, and complement, serving as a source of chronic inflammation. Oxidative stress, lipid metabolism abnormalities, and activation of the innate immune system are complexly involved, disrupting the homeostasis of the RPE-Bruch membrane-choriocapillaris complex. Subsequently, the pathway diverges into two branches.
Aged RPE cells exhibit a senescence-associated secretory phenotype (SASP), with increased expression of SA-β-gal, p53, p21, and p167). This senescent RPE phenotype is consistent with findings in patients with atrophic AMD, and selective removal by senolytics is attracting attention as a therapeutic target.
Genetic polymorphisms in CFH, C3, and ARMS2 are associated with susceptibility to AMD2). Dysregulation of complement pathways (classical, alternative, and lectin pathways) drives the expansion of geographic atrophy3). Inhibiting complement C3 is expected to suppress the entire terminal pathway downstream of C5, and inhibiting C5 is expected to prevent the formation of the membrane attack complex (MAC).
Anegondi et al. (2025) analyzed data from lampalizumab trials and showed that faster growth rates of geographic atrophy lead to faster BCVA decline, with approximately 75% losing ≥5 letters, 50% losing ≥10 letters, and 25% losing ≥15 letters over 2 years3).
Pachychoroid is a condition characterized by dilation of large choroidal vessels (pachyvessels) and increased choroidal vascular permeability1). Central serous chorioretinopathy (CSC) is a representative pachychoroid disease, and MNV arising from CSC or pachychoroid pigment epitheliopathy (PPE) is called pachychoroid neovasculopathy (PNV). The CFH gene has also been reported to be involved in the development of pachychoroid and CSC1).
Ribatti et al. (2024) reported that tryptase secreted by choroidal mast cells degrades Bruch’s membrane, inducing RPE death, and the release of VEGF-A, FGF-2, IL-8, and NGF promotes angiogenesis5).
Non-exudative MNV is biologically active even when asymptomatic, with persistent area enlargement4).
Wang et al. (2023) reported in a study of 45 eyes using SS-OCTA that growing MNV (area increase ≥50%) had a significantly shorter time to exudative conversion than non-growing MNV (13.60 months vs. 31.11 months, HR 12.51), and smoking history and hypertriglyceridemia were significantly correlated with growth (P=0.021)4).
Geographic atrophy often begins around the fovea, and the fovea shows relative resistance to atrophy 3). This produces the horseshoe-shaped or ring-shaped GA. The median time to foveal involvement is 2.5 years, during which high-contrast visual acuity is preserved, but daily visual functions such as dark-adapted sensitivity and reading speed are impaired early 8).
In 2023, two complement inhibitors for geographic atrophy received FDA approval in the United States.
However, neither drug achieved statistical significance in trials with visual acuity improvement as a prespecified endpoint, leaving a “structure-function dissociation” as a challenge 3)8). The European Medicines Agency (EMA) required demonstration of functional benefit and did not approve pegcetacoplan 8).
The complement factor D inhibitor lampalizumab failed to show suppression of GA area in the Phase III CHROMA (906 patients) and SPECTRI (975 patients) trials, leading to discontinuation. The visual cycle inhibitor emixustat also failed to show efficacy in the SEATTLE trial (580 patients).
In a review by Dinah et al. (2025), BCVA was found insufficient to capture the functional impact of geographic atrophy, and microperimetry, low-luminance visual acuity (LLVA), and reading speed were recommended as alternative endpoints 8). Establishing a composite, multimodal functional assessment will be key for future clinical trials.
Ji et al. (2025) reported a case of dry AMD treated with photobiomodulation (red to near-infrared light 650–1,300 nm) 6). Over 8 months, drusen area decreased by 58% in the right eye and disappeared completely in the left eye. Visual acuity improved from 20/30 to 20/20 in both eyes. The Lightsite III phase 3 trial also confirmed drusen volume reduction and visual acuity improvement.
Chung & Kim (2022) reported that the MDM2 inhibitor Nutlin-3a is a promising novel approach to selectively eliminate senescent RPE cells 7). Development of mitochondria-specific senolytics remains a future challenge.
Stem cell therapy using RPE cell transplantation is in the research stage, with multiple trials ongoing. Gene therapy targeting complement factors is also being investigated.
Genetic polymorphisms in CFH, ARMS2, and C3 are associated with the risk of developing AMD. Genotyping may enable prognosis prediction, such as the development of AMD in the fellow eye 1). However, at present, treatment strategies based on genetic testing are not standardized 2), and routine testing is not recommended.
Atrophic AMD progresses slowly, but when geographic atrophy reaches the fovea, visual acuity drops to 0.1 or less. The median time to legal blindness (ETDRS <20 letters) is reported to be 6.2 years 3). Analysis of lampalizumab trial data showed that mean BCVA decreased from 66 to 57 letters (approximately 20/50 to 20/80 equivalent) over 2 years 3).
Faster growth of geographic atrophy is associated with faster BCVA decline; eyes with a single subfoveal lesion in the fastest growth group lost about 4 lines (17.75 letters) over 2 years 3). In contrast, the slowest growth group lost only 1.69 letters over 2 years. Even when the fovea is preserved, scotopic sensitivity, contrast sensitivity, and reading speed are impaired early, so visual acuity alone underestimates functional impact 8).
Additionally, there is a certain probability of developing MNV and transitioning to neovascular AMD, which can lead to more severe vision loss.
The visual prognosis has significantly improved with the control of MNV using anti-VEGF drugs. In the 5-year follow-up of the CATT study, 50% of eyes achieved visual acuity of 20/40 or better 11). However, if left untreated, approximately 90% of eyes deteriorate to visual acuity of 0.1 or less, leaving fibrotic or atrophic scars in the macula. In cases with massive hemorrhage from MNV, extensive visual field defects may occur, leading to more severe visual dysfunction including complete blindness.
Complete cure of MNV is not possible, and without appropriate treatment and long-term management, irreversible vision loss can easily occur 1). Even if MNV activity temporarily subsides, it may recur over the long term, and repeated exudation can lead to atrophic changes and fibrotic scarring. Considering that MNV also occurs at a high rate in the fellow eye, continued treatment and regular monitoring are essential 1). Active low vision care is recommended for patients with severe visual impairment.
日本網膜硝子体学会 新生血管型加齢黄斑変性診療ガイドライン作成ワーキンググループ. 新生血管型加齢黄斑変性の診療ガイドライン. 日眼会誌. 2024;128(9):680-696.
Flaxel CJ, Adelman RA, Bailey ST, et al. Age-Related Macular Degeneration Preferred Practice Pattern. Ophthalmology. 2024;131(8):S1-S50.
Anegondi N, Steffen V, Sadda SR, et al. Visual loss in geographic atrophy: learnings from the lampalizumab trials. Ophthalmology. 2025;132(4):420-430.
Wang Y, Sun J, Wu J, et al. Growth of nonexudative macular neovascularization in age-related macular degeneration: an indicator of biological lesion activity. Eye (Lond). 2023;37(10):2048-2054.
Ribatti D, Dammacco R. Mast cells in human choroid and their role in age-related macular degeneration (AMD). Clin Exp Med. 2024;24(1):98.
Ji PX, Pickel L, Berger AR, Sivachandran N. Improvement in dry age-related macular degeneration with photobiomodulation. Case Rep Ophthalmol. 2025;16(1):155-162.
Chung H, Kim C. Nutlin-3a for age-related macular degeneration. Aging (Albany NY). 2022;14(14):5613-5616.
Dinah C, Esmaeelpour M, Rachitskaya AV, De Salvo G, Munk MR. Functional endpoints in patients with geographic atrophy: what to consider when designing a clinical trial. Surv Ophthalmol. 2025. doi:10.1016/j.survophthal.2025.02.004.
Heier JS, Lad EM, Holz FG, et al. Pegcetacoplan for the treatment of geographic atrophy secondary to age-related macular degeneration (OAKS and DERBY): two multicentre, randomised, double-masked, sham-controlled, phase 3 trials. Lancet. 2023;402:1434-1448.
Khanani AM, Kotecha A, Chang A, et al. TENAYA and LUCERNE: two-year results from the phase 3 neovascular age-related macular degeneration trials of faricimab. Ophthalmology. 2024;131:914-926.
Martin DF, Maguire MG, Ying GS, et al; CATT Research Group. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;364:1897-1908.
