Stargardt Disease

Etiology

STGD (Online Mendelian Inheritance in Man, OMIM #248200; juvenile macular degeneration; macular dystrophy with flecks, type 1; fundus flavimaculatus; or early-onset severe retinal dystrophy) is commonly inherited as an autosomal recessive trait due to missense and null mutations in the adenosine triphosphate-binding cassette transporter alpha 4 subunit (ABCA4) gene.[4] The ABCA4 gene is complex and consists of 50 exons mapped to the short arm of the human chromosome 1. While the former is associated with a milder, adult-onset, "foveal-sparing" disease, the latter can lead to severe and early-onset disease. There are few exceptions to this rule as a severe disease secondary to a few missense variants like p.Leu541Pro, p.Ala1038Val complex, and p.Arg1640Trp have been noted.[1] Other diseases associated with the mutation of the ABCA4 gene include cone-rod dystrophy 3 (OMIM #604116, autosomal recessive), retinitis pigmentosa 19 (OMIM #601718, autosomal recessive), and age-related macular degeneration 2 (OMIM #153800, autosomal dominant). ABCA4-associated retinopathies or ABCA4 retinopathies encompass a group of monogenic autosomal recessive diseases caused by varying amounts of decreased ABCA4 function.[5][6] The spectrum of phenotypes includes bull's eye maculopathy, cone-rod dystrophy, retinitis pigmentosa, rapid-onset chorioretinopathy,[7] and deep scleral exposure due to extensive chorioretinal damage.[6][8] It is thought that the progressive reduction of function of the ABCA4 gene produces Stargardt disease, cone-rod dystrophy, and retinitis pigmentosa, respectively.[5] Retinitis pigmentosa has the most severe phenotype with a flat electroretinogram and absent ABCA4 function.[5] 

Stargardt-like diseases are retinal diseases with phenotypes similar to STGD1 but with an autosomal dominant inheritance.[9][10] Such diseases include STGD2, STGD3, and STGD4. STGD2 was erroneously linked to chromosome 13q34[11]; however, this phenotype occurs due to mutation in the ELOVL4 gene at 6q14.1 as in STGD3 (OMIM #600110, Stargardt disease type 3; macular dystrophy with flakes, type 3; autosomal dominant Stargardt-like macular dystrophy).[12] Patients with STGD3 do not usually show the silent or jet-black dark choroid as seen in STGD1; therefore, fluorescein angiograms may be darker than usual.[9][13][14] Variants in ELOVL4 are associated with autosomal dominant spinocerebellar ataxia, autosomal recessive disorder congenital ichthyosis, spastic quadriplegia, and impaired intellectual development (ISQMR). 

STGD4 (OMIM #603786, Stargardt disease type 4) is an autosomal dominant disease caused by the prominin-1 gene (PROM1) mutation at chromosome 4p15.32.[10][15] The disease severity may increase in such cases if an additional mutation in ABCA4 is also present.[16] PROM1 mutations can also cause macular dystrophy, bull's eye maculopathy, cone-rod dystrophy, and retinitis pigmentosa.[6] WDR19 (IFT144 or NPHP13, chromosome 4p14), a gene classically associated with ciliopathies including Senior-Løken syndrome, Cranio-ectodermal dysplasia, nephronophthisis, short-rib thoracic dysplasia 5 with or without polydactyly, may also cause Stargardt disease-like phenotype.[17]

Other genes that can cause similar phenotypes include PRPH2 (multifocal pattern dystrophy simulating fundus flavimaculatus), CDH3, CERKL, CRX, BEST1, IMPG1, RDH12, and RPGR;[17] however, all these phenotypes do not show the typical triad of Stargardt disease.

Epidemiology

Although Stugardt disease is the most common cause of juvenile macular dystrophy, its prevalence in the United States is estimated to be 10 to 12.5 per 100,000 individuals.[2] Cornish et al. estimated the annual incidence of this condition in the United Kingdom to be around 0.110 to 0.128 per 100,000 individuals.[18] A study conducted in the Netherlands found the incidence of Stargardt disease to be 1.67 to 1.95 per 1,000,000 individuals per year.[19] The point prevalence of Stargardt disease was 1:22,000 to 1:19,000 in 2018 in the Netherlands.[19] 

Pathophysiology

The ABCA4 gene encodes for the rim protein (RmP), a retinal adenosine triphosphate (ATP)-binding cassette protein. RmP is present on the membrane of the discs in the outer segment of the photoreceptors (rods and cones),[20] and participates in the visual cycle-related vitamin A metabolism. The light photons lead to isomerization of the 11-cis retinal to the all-trans retinal form. The all-trans-retinal form then enters the intradiscal space, where some of its molecules combine with a phospholipid (phosphatidylethanolamine, PE) ordinarily present in the disc membranes to form a complex called N-retinylidene-PE (N-RPE).

The RmP protein breaks the N-RPE complex and transports it from the intradiscal space. The absence of the RmP protein leads to the accumulation of the N-RPE complexes, leading to N-retinylidene-N-retinyl-PE (A2PE) formation by combining N-RPE and all-trans-retinal. A2PE hydrolyzes to form N-retinylidene-N-retinyl-ethanolamine (A2E). The A2E produced during these reactions leads to lipofuscin formation, accumulating in the retinal pigment epithelium (RPE) cells. The accumulation of lipofuscin leads to the formation of flecks. In vivo, study results have shown that patients with STGD1 have 2 to 5 times the amount of lipofuscin compared to the age-matched controls.[21][22]

Lipofuscin is resistant to lysosomal enzymatic degradation. High levels of A2E slow the phagocytic ability of the RPE cells, ultimately leading to RPE and photoreceptor cell death.[23][24] 

Multimodal imaging studies have suggested that the degeneration of the photoreceptors’ inner and outer segments leads to hyper-reflective debris on the apical surface of the RPE cells, seen clinically as flecks. These flecks cause gradual disruption of the interdigitation zone (IZ), the ellipsoid zones (EZ), the external limiting membrane (ELM), and, eventually, the thinning of the outer nuclear layer (ONL).[25]

Histopathology

Light microscopy studies have shown the presence of abnormal RPE cells and loss of photoreceptors and RPE cells in eyes with STGD1. In addition, the RPE cells are heterogeneous in size and contain numerous PAS-positive intracytoplasmic material, lipofuscin.[26] Reactive Müller cell hypertrophy may be noted, which stains positively for glial fibrillary acid protein.[27] Cones are more affected than rod cells in all retinal locations, and the macula may show severe degeneration.[26]

History and Physical

Patients with STGD1 usually present with progressive, painless bilateral central vision loss. The presenting visual acuity ranges from 20/20 to 20/400. The other presenting symptoms include central scotoma, delayed dark adaptation, photosensitivity, photopsia, and abnormal color vision. Usually, both eyes are involved. Both males and females can be affected. The age of presentation is usually childhood, with a second peak at early adulthood.[2] The early-onset disease may cause severe vision loss compared to late-onset or adult-onset disease. A nonsense mutation usually causes the early-onset disease and severe loss of function of ABCA4, and late-onset "foveal-sparing" Stargardt disease (FS-STGD1) can be seen in patients with less reduction of ABCA4 function (missense mutation).[2]

There is a marked phenotypic heterogeneity of STGD1 due to a considerable number (more than 1200) of disease-causing mutations of varying severity and different types in ABCA4.[2][28] The phenotypes vary in age of onset, severity, visual acuity, multimodal imaging features, and electrophysiological characteristics. The typical triad associated with STGD1 includes flecks, macular atrophy, and peripapillary sparing.[29] Another crucial feature is jet-black dark choroid or silent choroid in the fluorescein angiogram.

Flecks are yellowish-white round or pisciform (fish-tail) shaped lesions of varying sizes located within the RPE. They tend to develop centrifugally from the macular region and may even progress beyond the vascular arcades to the mid-periphery but usually spare the far periphery. Around 30% of patients with childhood-onset disease may not show flecks, which may appear as the child grows.[30] This is a critical cause of misdiagnosis.

The macula is involved early in the disease course. The macular atrophy changes are usually progressive. However, its severity can be variable, ranging from a small region of central macular atrophy to atrophy beyond the vascular arcades. In a child with STGD1 with visual decline, the macular changes may be very subtle and missed on routine ophthalmoscopic examination. The severity of visual decline may not correlate with a normal-appearing retina. Multimodal imaging, including autofluorescence, optical coherence tomography, and electroretinography, helps to diagnose such cases. The macular manifestations include normal macula, vermillion fundus, abnormal foveal reflex, mild retinal pigment epithelial changes, pigmentary changes in the macula, foveal thinning, beaten bronze metal appearance, snail-slime appearance, bull's eye macula (BEM), dense pigmentation at the macula, macular atrophy, and even scleral show with very severe chorioretinal atrophy that may extend beyond the arcades.[8]

Peripapillary sparing (absence of flecks around the optic disc) is a diagnostic feature in STGD1 that is more prominently visible on fundus autofluorescence imaging. Pentosan polysulfate maculopathy, which is a close differential diagnosis of STGD1, may not spare the peripapillary region.[31]

BEM may be seen in STGD1. The differential diagnoses for BEM include hydroxychloroquine/chloroquine retinopathy, STGD1, cone dystrophy, cone-rod dystrophy, advanced dry age-related macular degeneration, central areolar choroidal dystrophy, benign concentric annular macular dystrophy, fenestrated sheen macular dystrophy, juvenile Batten disease, fucosidosis, ceroid lipofuscinosis, and clofazimine maculopathy. Flecks tend to appear later in eyes with this phenotype.

Chorioretinal degeneration is the most severe phenotype related to ABCA4 gene mutation. Choriocapillaris atrophy, widespread nummular pigmentary deposits, and bony spicule characterize ABCA4 gene mutation. Severe cases can show a visible sclera, vascular attenuation, and have peripapillary disease involvement. Florid retinitis pigmentosa with bony spicule pigmentation may be a severe form of ABCA4 retinopathy.[5]

Foveal-sparing Stargardt disease (FS-STGD1) is a milder phenotype that presents late and has an excellent visual prognosis with foveal sparing. The usual age of presentation is 35 to 45 years.[32] Such cases may be misdiagnosed as age-related macular degeneration. The central visual acuity is usually preserved (at least 20/40), though the rate of progression may vary due to the phenotypic variability associated with ABCA4.[33] On autofluorescence imaging, such patients have normal foveal autofluorescence surrounded by at least 180 degrees of retinal atrophy (hypoautofluorescence).[33] Factors leading to foveal sparing and relatively better prognosis include genetic factors (less severe missense mutation or hypomorphic mutation), environmental factors, the individual's cone density, and macular pigment density.[2]

Evaluation

The various imaging modalities have provided great insights into the anatomical and functional status of the retina. They are valuable for screening and early diagnosis of the disease. Lambertus et al reported that almost one-fourth of patients diagnosed with childhood-onset STGD1 had no clinically identifiable retinal lesions at presentation.[34] Patients can easily be misdiagnosed and subjected to multiple investigations like neuroimaging, lumbar punctures, and therapies like occlusion. Multimodal imaging can help in early and correct diagnosing, monitoring the disease progression, and evaluating the treatment modalities' efficacy. The emergence of novel ABCA4-specific molecular and genetic therapies can help patients with the early stages of the disease. 

i) Color fundus photography (Figure 1. ) is helpful in the documentation and monitoring of the type of retinal changes and extent of the disease. The Fishman classification is based on color fundus photographs.[35] Stage 1 is characterized by macular pigmentary changes, which may range from faint or irregular pigment mottling to bronze-beaten appearance; a ring of pisciform flecks within 1 disc-diameter (DD) on all sides of the fovea; and a normal electroretinogram (ERG) and electrooculogram (EOG). The presence of pisciform flecks beyond 1 DD of the fovea and may extend beyond the vascular arcades and nasal to the optic nerve; a normal ERG and EOG except for occasional subnormal cone/rod response; and prolonged dark adaptation is representative of stage 2. Atrophic macular changes replacing the previously recognized flecks, subnormal EOG and ERG responses, and central and mid-peripheral and peripheral visual field defects characterize stage 3. Stage 4 occurs with extensive atrophy throughout the entire fundus, reduced ERG responses, and moderate to extensive peripheral field restriction.[35]

Noble and Carr classified STGD1 (Stargatdt disease and fundus flavimaculatus) based on fundus appearance into 4 groups:[36]

• Group 1: macular degeneration without flecks
• Group 2: perifoveal flecks with macular degeneration (Stargatdt disease)
• Group 3: diffuse flecks with macular degeneration (fundus flavimaculatus with macular degeneration)
• Group 4: diffuse flecks, absent macular degeneration (pure fundus flavimaculatus)

ii) Fundus fluorescein angiography:[14] The hallmark finding is the "dark choroid," which appears as a highlighted retinal blood vessel against a hypofluorescent choroid. The dark choroid is a diagnostic feature seen in at least 80% of patients. Results from a study found the "dark choroid sign" in 94% of patients with STGD1 and ABCA4 mutation[37] occurring due to blockage of background choroidal fluorescence—because of lipofuscin in the RPE cells. Flecks appear hypofluorescent due to the blocked transmission but may appear hyperfluorescent in the presence of atrophy. They can be seen on fundus fluorescin angiography (FFA) before they become clinically visible. The regions of RPE atrophy appear hyperfluorescent due to transmission defects, and a dark choroid ring may be seen.[37]

iii) Infrared imaging: Flecks typically appear hyperreflective; however, the resorbing flecks appear hyporeflective. The macular atrophy region also appears hyperreflective and is usually surrounded by a dark ring.[37]

iv) Optical coherence tomography: Optical coherence tomography (OCT) findings include thinning outer retinal layers and disruption of the EZ. The flecks appear as subretinal and intraretinal hyperreflective deposits.

Querques et al categorized flecks into 2 types.[38] While dome-shaped hyperreflective deposits characterize type 1 flecks on the inner part of the RPE, type 2 flecks represent small linear deposits in the ONL and spare the RPE. Voigt and colleagues have devised a newer classification describing 5 subtypes of flecks.[38] Class A flecks are seen as hyperreflective deposits at the level of the outer segment of the photoreceptors, RPE, interdigitation zone (IZ), and RPE-Brunch membrane complex. The other subtypes progress toward the inner retinal layers. Class B flecks reach the ELM, and Class C flecks break into the ONL. Class A to C represent the type 1 flecks per the previous classification. Class D flecks are only present within the ONL, while deposits in the outer retinal layers disappear; these only represent the type 2 flecks as per the previous classification. Class E flecks represent atypical drusen-like lesions.[38] 

The other OCT changes noted previously to the fundoscopy changes are visible, including a subfoveal optical gap and thickening of the ELM or hyper-reflectivity at the base of ONL.[39][40][41] The disruption of the EZ and thinning of proximal ONL characterizes the subfoveal optical gap.

Noupuu et al described the progression of this optical gap in 3 stages:

• Stage 1 represents the mild disruption of the EZ in the foveal region
• Stage 2 represents additional photoreceptor loss  
• Stage 3 represents the involvement of even the inner retinal layers[39] 

Earliest changes are seen in the parafoveal region and not at the central fovea.[42] OCT appears more sensitive in detecting early lesions compared to fundus autofluorescence.[43]

Cai et al demonstrated that the total area of EZ loss gradually expanded at around 0.31 mm/year.[44] ELM thickening may represent an early sign in pediatric patients. While Burke et al proposed that ELM thickening could signify a Muller cell response,[40] Pang et al suggested that it could represent the migration and retraction of the inner segment of the photoreceptors.[41]

Multiple studies have used enhanced depth imaging (EDI) OCT and swept-source (SS)-OCT to study the choroidal thickness in STGD1 patients. Many studies have revealed that the choroid thickness reduces, affecting mainly the tiny choroidal vessels. Similarly, there is a significant reduction in the choroidal vascularity index. So far, there are 4 patterns of choroidal morphology described. Pattern 1 is characterized by a normal-appearing choroid, pattern 2 by a reduction of the thickness of either the Sattler or Haller layer, pattern 3 by a reduction of the thickness of both Sattler and Haller layers, and pattern 4 by a reduction of both Sattler and Haller layers with choroidal caverns. These patterns positively correlated with degenerative retinal changes on OCT and signs of progression.[45]

Other than thickness, hyperreflective foci in the choroid are also noted. Piri et al demonstrated that the number of hyperreflective foci in the Bruch membrane/RPE complex, choriocapillaris, and Sattler layer correlated with increasing disease severity in terms of central macular thickness, visual acuity, and duration of the disease.[46]

v) OCT-angiography: Study results have demonstrated reduced superficial and deep vessel densities and choriocapillaris vessel density attenuation. OCT-angiography may also help diagnose choroidal neovascular membrane (CNVM), which is a rare and late complication

vi) Fundus autofluorescence: Monitoring the disease progression uses both short-wavelength fundus autofluorescence (SW-FAF) and near infrared-wavelength fundus autofluorescence (NIR-FAF). Flecks are seen on SW-FAF as hyper AF (Figure 2. ), while they appear hypo AF on NIR-FAF. Flecks can be seen on NIR-FAF before they appear on SW-FAF. The atrophic areas show decreased AF on both types of FAF, with the area appearing larger on NIR-FAF than on SW-FAF, which may be due to lipofuscin released by the degenerating photoreceptors, producing a falsely typical image despite the presence of multiple atrophic RPE lesions. Hyper AF lesions on SW-FAF are usually the earliest features and may appear before flecks become visible. The RPE loss expands at a rate of 0.28 to 1.58 mm/year.[44]

Fujinami et al classified the AF appearance on SW-AF images into 3 types. Localized low foveal AF signal surrounded by a homogenous AF background characterizes type 1. Type 2 appears as localized low macular AF surrounded by a homogenous AF background with multiple areas of abnormal signal. Multiple areas of low AF signal at the posterior pole and a heterogenous background FAF are identifiable in type 3.[47] The rate of atrophy enlargement (RAE) in each of the 3 types was estimated to be 0.06, 0.67, and 4.37 mm2/year, respectively.[47]

Kuehlewein and colleagues have divided the lesions into definitely decreased AF (DDAF) and questionably decreased AF (QDAF). Lesions with darkness ≥ 90% of the darkness of the optic nerve head (ONH) are described in DDAF; QDAF has lesions with a darkness level of 50% to 90% of the ONH.[48]

The ProgStart study report number 5 revealed that around 50% of the eyes that did not present with DDAF developed the lesion in less than 5 years (median 4.93 years).[49] Klufas et al described a classification for the ultra-widefield (UWF) FAF.[50] Lesions isolated only to the macular region are in type 1, an atrophic macula, and peripheral flecks identify type 2, while type 3 appears as an atrophic macula with peripheral atrophy.[50]

Quantitative AF (qAF) was first described by Delori et al.[51] This classification is a standardized approach for assessing the number of lipofuscin levels in the retina. The qAF levels in STGD1 are raised, even without fundus abnormalities; therefore, the classification helps establish the diagnosis in borderline cases and monitor the efficacy of the various therapeutic interventions.

vii) Adaptive optics scanning laser ophthalmoscope: Observed are areas with increased spacing between the photoreceptor (both rods and cones). Findings from adaptive optics scanning laser ophthalmoscopy (AOSLO) suggest that photoreceptor degeneration is the primary pathology rather than an event secondary to RPE cell death. Song et al further demonstrated the centrifugal progression of the disease.[52]

viii) Visual field:[53] This test shows a progressive central scotoma. Loss of mid and peripheral VFs is usually associated with more severe disease. Schroeder et al classified the VF severity into 3 groups. Group 1 has a central scotoma within 10°, group 2 has central scotomas within 10° to 35°, and group 3 only has "temporal residues."[54]

ix) Microperimetry: The ProgStar study is a reliable investigation for detecting macular function in studies evaluating the various treatment modalities. The study results showed that measuring visual acuity is not a sensitive outcome for monitoring the disease in various treatment trials. VF testing can miss small scotomas. Patients with STGD1 show varying degrees of reduced sensitivity on microperimetry (MP), depending on the stage of the disease. Macular sensitivity is reduced in the areas with flecks and abnormal AF.

The ProgStar study results also found the mean sensitivity reduces by around 0.68 dB yearly.[55] The "Scotopic Microperimetric Assessment of Rod function in Stargardt Disease" (SMART) study, the ancillary study to ProgStart, demonstrated that the loss of scotopic macular function in the extrafoveal spots was faster than the loss of mesopic macular function.[56]

 x) Visual electrophysiology: Lois and colleagues classified STGD1 cases into 3 groups based on their electrophysiological properties.[57] 

• Group 1 shows severe abnormalities in pattern electroretinogram (PERG) with normal photopic and scotopic full-field electroretinogram (ffERG).
• Group 2 includes additional abnormalities (loss) in photopic ffERG (abnormal cone function).
• Group 3 shows loss of both photopic and scotopic full-field ERG response (abnormal cone and rod function).[57][58][57]

Fujinami and colleagues noted that patients with initial rod involvement in ERG had clinically significant electrophysiological deterioration with time (mean follow-up of around 10 years).[58] About 22% of group 1 patients showed clinically significant progression with time.[58] Multifocal ERG can detect foveal dysfunction, and the area of dysfunction may be larger than the area of morphological changes.[59] Electrooculogram is affected variably and is more involved in eyes with flecks (69%) than eyes without flecks (42.5%).[60]

xi) Genetic testing: Genetic testing is vital as various changes noted in the ABCA4 gene can cause a variety of phenotypes. Genetic testing also helps in genetic counseling, prognostication, and preconception planning.

Treatment / Management

Currently, no treatment modality is recommended by the FDA to prevent or reverse visual loss in patients with STGD1. Patients should use photoprotection to delay the disease progression and low-vision aids for those with visual field loss. Prescribe patients with proper refractive correction and advise patients to avoid smoking. Intravitreal anti-vascular endothelial growth factor (VEGF) injections are the preferred treatment modality for patients who develop CNVM. However, several ongoing clinical trials are investigating various novel therapeutic modalities for this condition.

Pharmacological Therapy

The investigation of several drugs that can reduce lipofuscin accumulation is currently underway. Reducing lipofuscin can be achieved by either inhibiting the retinoid cycle by inhibiting the key enzymes (emixustat and isotretinoin) or disrupting the transporters like retinol-binding protein 4 (RBP4) to decrease the toxic bisretinoid formation (fenretinide, A1120, BPN-14136, STG-001, LBS-008 or tinlarebant).[61] These drugs include the following:

• Emixustat hydrochloride (ACU-4429): inhibits the enzyme isomerase RPE65 (all-trans: 11-cis retinol isomerase) in a dose-dependent reversible manner. ACU-4429 prevents the isomerization of all-trans-retinyl ester to 11-cis-retinol, thus reducing the production of toxic by-products like A2E. A Phase 1a study demonstrated a dose-dependent reversible suppression (returned to the baseline by day 7) of the enzyme, which manifested as the slow recovery of the rod b-wave amplitude after photobleaching on ERG.[62] A Phase 1b study revealed that the drug was systematically safe but had few reversible ocular adverse effects like chromatopsia, blurred vision, and reduced visual acuity.[63] A phase 2 clinical trial studied the effect of 3 doses (2.5 mg, 5 mg, or 10 mg) of daily emixustat for 1 month. The results showed that 10 mg, 5mg, and 2.5 mg doses resulted in near-complete (mean = 91.86%, median = 96.69%), moderate (mean = 52.2%, median = 68.0%), and no suppression (mean = −3.31%, median= −12.23%) of the rod-derived b-wave amplitude recovery rate on ERG after photobleaching respectively.[64] 
(A1)

• ALK-001 (deuterated vitamin A): ALK-001 slows the rate of vitamin A dimerization, decelerating the rate of A2E and lipofuscin formation. Results from studies evaluating its efficacy in murine models have shown promising results with no adverse effects on retinal functions.[65] A phase 2 clinical trial evaluated the drug’s pharmacokinetics and 1-year interim safety in 50 subjects aged ≥ 18 suffering from ABCA4-related STGD1 and had a well-delineated area of atrophy. Results of the trial showed that it was well-tolerated, and there were no adverse reactions such as night blindness, impaired dark adaptation, and clinically significant derangement of liver function tests or serum vitamin A levels. An expanded study is underway to recruit new participants ≥ 12 years.

• Fenretinide (retinoid-based RBP4 antagonist): Fenretinide is a synthetic derivative of vitamin A and is a competitive inhibitor of retinol, as it prevents retinol binding to the retinal binding protein (RBP) in the eye. The RBP-fenretinide complex is excreted in the urine, thus reducing the circulating RBP levels; this significantly reduces the growth of geographic atrophy and the onset of CNVM in patients with dry age-related macular degeneration (ARMD). However, there are concerns regarding its long-term safety due to reports of teratogenicity and hemangiosarcomas formation in mice.[66]
(A1)

• Vitamin A: Per the current recommendations, avoid vitamin A dietary supplements, as excess vitamin A is considered detrimental. However, these recommendations have a theoretical basis only. To date, there is little evidence to support or exclude the role of vitamin A in deteriorating the disease.[67]

Gene Therapy

The goal is to introduce a functional ABCA4 gene that can produce an adequate amount of the standard, active transporter protein in the photoreceptor cells, thus preventing disease progression. Adeno-associated viruses (AAV) deliver genes to the ocular tissues effectively. However, the AAV vector’s limited packaging capacity (4.7 kb) renders it insufficient to pack the large ABCA4 gene (7 kb). The help of a “dual-vector strategy,” ie, large genes split in half and packaged in 2 separate AAV vectors, can solve this problem.[68]

Recently developed lentivirus-based vectors deliver large genes. One such lentivirus-based vector is delivered subretinally.[69] The gene can be delivered via deoxyribonucleic acid (DNA) nanoparticles and DNA-encapsulating liposomes. An intravitreal or subretinal injection can accomplish delivery.[61](B3)

Stem-cell therapy: This therapy can treat the disease by regenerating the RPE cells.[61]

Genetic counseling: Genetic counseling is an integral part of managing this condition. Genes can help predict the prognosis and progression of the disease (ie, STGD1 is autosomal recessive, STGD3 and STGD4 are autosomal dominant).

Differential Diagnosis

Several inherited macular dystrophies have a clinical appearance similar to STGD1. Distinguishing STGD1 among these similar-appearing conditions is essential for prognostication, systemic associations, and genetic counseling.[70]

Pattern Dystrophy 

Marmor and Byers were the first to propose “pattern dystrophy” when describing a spectrum of characteristic fundus patterns (ie, butterfly, reticular pigmentary). Pattern dystrophy (PD) most commonly masquerades as STGD1.[71] PD is an autosomal dominant condition with mild or no visual loss in the second to fifth decade. PD associated with the PRPH2 gene (multifocal pattern dystrophy simulating fundus flavimaculatus) primarily manifests as a macular disease. However, the phenotype may range from macular flecks to diffuse pigmentation in the periphery. Peripapillary sparing is a characterization of PD similar to STGD1. However, PD is associated with a higher incidence of CNVM. FFA helps distinguish the two as there will be window defects in PD, in contrast to a “masked choroid” in STGD1.[72]

Autosomal Dominant Stargardt-like Macular Dystrophies

These dystrophies include variants associated with the ELOVL4 (STGD2 or STGD3) and the PROM1 (STGD4) genes.[70]

Batten Disease/Juvenile Neuronal Ceroid Lipofuscinosis 

Defects in the processing of lysosomal storage characterize Batten disease/juvenile neuronal ceroid lipofuscinosis, a rare metabolic disease. These defects lead to apoptosis of both the photoreceptor (retina) and the neuronal (brain) cells. Visual loss and severe color vision deficiency are common, and the life expectancy of patients with this condition is usually 20 to 30 years.

The disease progression is usually rapid. Ocular examination shows vascular attenuation and optic disc pallor. OCT reveals a thinned retinal nerve fiber layer, while dark-adapted full-field ERG shows an absent or electronegative wave. Maintain high suspicion in younger patients diagnosed with severe STGD1 with rapid progression.[73]

Maternally Inherited Diabetes and Deafness or Other Mitochondrial Diseases (Including Mitochondrial Encephalomyopathy, Lactic Acidosis, and Stroke-Like Episodes)

Patients with maternally inherited diabetes and deafness (MIDD) or other mitochondrial diseases present in the third to seventh decade. Ocular examination shows pigmentary changes at the macula, areas of RPE atrophy, and fleck-like deposits.[74] FAF shows hyperautofluorescent spots with corresponding subretinal deposits on OCT. Patients with this condition have distinguishing features such as personal or family history of early-onset diabetes mellitus and hearing loss. Other distinguishing features include the absence of a dark choroid on FFA and the presence of outer retinal tubulations on OCT.[75]

Age-Related Macular Degeneration 

This condition can mimic late-onset STGD1, as the flecks may be mistaken for drusens. A careful examination can help distinguish the two, as drusen tend to become confluent toward the center of the macula. Flecks show intense hyper AF, while drusens may be mildly hyperautofluorescent or hypoautofluorescent.[76] Sub-RPE accumulation on OCT characterizes drusens, while STGD1 is associated with hyperreflective thickening of the RPE. Age-related macular degeneration does not show a dark choroid on FFA.

Pentosan Polysulfate Maculopathy 

This condition is associated with the intake of pentosan, used in treating interstitial cystitis. The FAF changes may not spare the peripapillary region in pentosan polysulfate maculopathy.[31]

Prognosis

Many patients with this condition progress to legal blindness from losing central vision, although peripheral vision tends to remain preserved. The prognosis is widely variable and is highly dependent on the age of onset of the disease; therefore, patients presenting with early onset have poor outcomes. Visual acuity in such patients declines rapidly, along with the appearance of progressive retinal degeneration—and can even lead to profound chorioretinal atrophy. Thus, patients with this phenotype have a severe disease with a poor prognosis.[34]

Rapid-onset chorioretinopathy (ROC) is a distinct sub-phenotype characterized by rapid disease progression.[7] Patients with late-onset disease maintain good visual acuity, even at an older age. The disease does not impact the patient’s general health or life expectancy.

Complications

The complications associated with STGD1 include the following:

Subretinal fibrosis: Minor ocular trauma can lead to subretinal fibrosis and RPE hypertrophy.[77][78] Grandinetti et al proposed that ocular trauma leads to the release of lipofuscin, which disrupts the RPE cells, leading to subretinal fibrosis and RPE hypertrophy.[77] (Gass et al presented a similar hypothesis.)[78] Rossi et al proposed that even microtrauma can lead to subretinal fibrosis as they found patients developing subretinal fibrosis without a distinct history of ocular trauma or contact sports.[79] Patients with STGD1 are advised to use eye protection while playing competitive sports.[78][79]

Choroidal neovascular membrane: This rare condition is a late complication of STGD1. 

Legal blindness: Many patients develop legal blindness with age.