Stargardt Disease 1

A number sign (#) is used with this entry because of evidence that Stargardt disease-1 (STGD1) is caused by homozygous or compound heterozygous mutation in the ABCA4 gene (601691) on chromosome 1p22.

Stargardt disease-3 (STGD3; 600110) is caused by mutation in the ELOVL4 gene (605512) on chromosome 6q14, and Stargardt disease-4 is caused by mutation in the PROM1 gene (604365) on chromosome 4.

A locus for Stargardt disease mapped to chromosome 13q34 and designated STGD2 was found to be in error; the disorder in the family in which the linkage was made was correctly mapped to chromosome 6q14 (STGD3).

One patient with a diagnosis of juvenile macular degeneration was found to be compound heterozygous for mutations in the CNGB3 gene (605080) on chromosome 8q.

Fundus flavimaculatus (FFM) is an allelic subtype of Stargardt disease that has been associated with mutation in the ABCA4 gene and the PRPH2 gene (179605). FFM has a later age of onset. If loss of visual acuity begins in the first 2 decades, the designation Stargardt disease is preferred; if it begins later in life and has a more progressive course, the term FFM is preferred (Weleber, 1994).

An early-onset severe form of retinal dystrophy (CORD3; 604116) is caused by homozygous null mutations in the ABCA4 gene.

Clinical Features

Stargardt disease is one of the most frequent causes of macular degeneration in childhood. It has onset between 7 and 12 years, a rapidly progressive course, and a poor final visual outcome. Although visual acuity is severely reduced, peripheral visual fields remain normal throughout life. Degeneration limited to the macular area of the retina was described in multiple sibs by Ford (1961) and by Walsh (1957).

Fundus flavimaculatus, which is a form of fleck fundus disease (see 228980), derives its name from the occurrence of many yellow spots rather uniformly distributed over the fundus. In some older patients the flecks fade with time as atrophy of the retinal pigment epithelium (RPE) increases. Round, linear, or pisciform lesions are distributed in the posterior pole, sometimes with extension to the equator, and with macular involvement. Network atrophy of the retinal pigment epithelium, and choroidal vascular atrophy are features. Central visual loss, loss of color vision, photophobia, paracentral scotoma, and slow dark adaptation are features. This is probably an autosomal recessive disorder. Klien and Krill (1967) observed a 'familial incidence...in 10 of 27 patients.' The 10 familial cases included 4 pairs of affected sibs with ostensibly normal parents who were, however, not examined in most instances. No parental consanguinity was described. In 1 instance the father and 2 daughters were affected. In the instance of an affected brother and sister, the father was black and the mother white.

Krill and Deutman (1972) concluded that recessive macular dystrophy was the disorder described and beautifully illustrated by Stargardt (1909), and also was the disorder that Franceschetti (1963) renamed fundus flavimaculatus. Krill and Deutman (1972) suggested the possibility of a rarer, phenotypically indistinguishable, autosomal dominant form. Hadden and Gass (1976) presented evidence that fundus flavimaculatus is the same as the Stargardt form of macular dystrophy.

Pearce (1975) reported 4 families with 9 affected persons. In 1 instance, 2 affected persons married and both of their children were affected. Carpel and Kalina (1975) described 3 affected sisters.

Isashiki and Ohba (1985) remarked on variable expression. Among the 3 children of normal first-cousin parents were a 12-year-old boy with bull's eye macular change and sparse fundus flavimaculatus type flecks, and an 11-year-old girl with numerous fleck lesions of FFM throughout the posterior fundus and virtually no macular change.

As pointed out by Weleber (1994), Rosehr (1954) found that 2 of the original patients described by Stargardt (1909), when seen almost 50 years later, still did not complain of night blindness and their visual fields were, at most, only mildly constricted. The macular regions showed marked atrophy in each patient, and 1 patient had peripheral pigment clumping and drusen.

Whereas Stargardt disease shows juvenile to young adult age of onset, the clinically similar retinal disorder fundus flavimaculatus often displays later age of onset and slower progression. Histologically the disorder is characterized by subretinal deposition of lipofuscin-like material. As pointed out by Meitinger (1997), Stargardt disease had always been considered to be a retinal degeneration originating in the retinal pigment epithelium, which underlies the photoreceptors, predominantly cones, of the macula. Thus, the findings of Allikmets et al. (1997) that it is a disease of the rods and that the particular mutant ABC transporter is expressed in rod photoreceptors but not in blue cones came as a surprise.

To understand better the shared characteristics of Stargardt macular dystrophy and fundus flavimaculatus, Armstrong et al. (1998) surveyed 52 patients with STGD and 48 patients with FFM over a period ranging from 1 to 22 years. They found that morphologic changes and retinal function degeneration were more severe in patients with FFM than in patients with STGD. The duration of the disease had a greater effect on patients with FFM than on patients with STGD.

Rotenstreich et al. (2003) reviewed the clinical findings in 361 patients with Stargardt disease. Eighty-two (23%) had visual acuity of 20/40 or better, whereas only 16 (4%) had acuity of worse than 20/400. The presence of foveal sparing on ophthalmoscopy was associated with a higher prevalence of 20/40 acuity or better. Survival analysis showed that the prognosis of patients who were seen initially with visual acuity of 20/40 or better was related to age at initial visit: the earlier the patient presented, the more rapidly the acuity was likely to decrease below 20/40.

Chen et al. (2011) studied the relationship between macular cone structure, fundus autofluorescence (AF), and visual function in 12 patients with Stargardt disease and 27 age-matched healthy individuals. Patients were 15 to 55 years old, and visual acuities ranged from 20/25 to 20/320. At least 1 disease-causing mutation in the ABCA4 gene was found in 11 of the patients. High-resolution images of the macula were obtained with adaptive optics scanning laser ophthalmoscopy (AOSLO) and spectral domain optical coherence tomography (SD-OCT). Central scotomas were present in all patients, although the fovea was spared in 3. The earliest cone spacing abnormalities were observed in regions of homogeneous AF, normal visual function, and normal outer retinal structure. Outer retinal structure and AF were most normal near the optic disc. Longitudinal studies showed progressive increases in AF followed by reduced AF associated with losses of visual sensitivity, outer retinal layers, and cones. Chen et al. (2011) concluded that their findings support a model of STGD disease progression in which lipofuscin accumulation results in homogeneously increased AF with cone spacing abnormalities, followed by heterogeneously increased AF with cone loss, and then reduced AF with cone and RPE cell death.

By AOSLO in a family in which 2 brothers had early-onset STGD1, Song et al. (2015) found increased cone and rod spacing in areas that appeared normal in conventional images. Cone loss predominated closer to the fovea with a greater contribution from rod loss in the periphery.

Fujinami et al. (2015) performed a retrospective study of 42 patients diagnosed with STGD in childhood at a single institution between January 2001 and January 2012. Median ages of onset and baseline examination were 8.5 and 12.0 years, respectively. Median baseline logarithm of the minimum angle of resolution visual acuity was 0.74. At baseline, 26 (67%) of 39 patients with available photographs had macular atrophy with macular/peripheral flecks; 11 (28%) has macular atrophy without flecks; 1 (2.5%) had numerous flecks without macular atrophy; and 1 (2.5%) had a normal fundus appearance. Flecks were not identified at baseline in 12 patients (31%). SD-OCT detected foveal outer retinal disruption in all 21 patients with available images. Electrophysiologic assessment demonstrated retinal dysfunction confined to the macula in 9 patients (35%), macular and generalized cone dysfunction in 1 patient (4%), and macular and generalized cone and rod dysfunction in 15 patients (60%). Fujinami et al. (2015) concluded that childhood-onset STGD is associated with severe visual loss, early morphologic changes, and often generalized retinal dysfunction, although patients typically have less severe fundus abnormalities on examination.

Lambertus et al. (2015) performed a retrospective study of 51 patients, aged 7 to 64 years, who were diagnosed with STGD at or younger than age 10 years. Among the patients were 37 with 2 or more ABCA4 mutations, 7 with 1 ABCA4 mutation and the presence of yellow-white flecks, and 7 not known to have an ABCA4 mutation but with yellow-white flecks and either a dark choroid or an atrophic macular lesion. The mean age at onset was 7.2 years (range, 1-10). The median times to develop best corrected visual acuity (BCVA) of 20/32, 20/80, 20/200, and 20/500 were 3, 5, 12, and 23 years, respectively. Initial ophthalmoscopy in 41 patients revealed either no abnormalities or foveal RPE changes in 10 and 9 patients, respectively; the other 22 patients had foveal atrophy, atrophic RPE lesions, and/or irregular yellow-white fundus flecks. On fluorescein angiography (FA), a dark choroid was found in 21 of 29 patients. On fundus autofluorescence (FAF), there was centrifugal expansion of disseminated atrophic spots, which progressed to eventual profound chorioretinal atrophy. SD-OCT revealed early photoreceptor damage followed by atrophy of the outer retina, RPE, and choroid. On full-field electroretinogram (ffERG) in 26 patients, 15 had normal amplitudes and 11 had reduced photopic and/or scotopic amplitudes at their first visit. No correlation between ffERG abnormalities and rate of visual loss was found. Thirteen of 25 patients had progressive ffERG abnormalities. Lambertus et al. (2015) concluded that early-onset STGD1 falls on the severe end of the spectrum of ABCA4-associated retinal phenotypes.

Parodi et al. (2015) performed a prospective study of 27 patients (54 eyes) with STGD1 to identify a correlation between near-infrared (NIR) and short-wavelength (SW) fundus autofluorescence (FAF) patterns within the foveal region and best corrected visual acuity (BCVA) values. Eyes showing a pattern of foveal hyper-FAF on NIR-FAF had a higher BCVA than eyes with a reduced FAF signal. Similarly, mean sensitivity within 2 degrees of the foveal region was significantly better in eyes with hyper-FAF than in eyes with hypo-FAF. Moreover, eyes with hyper-FAF on SW-FAF did not present a significant difference in BCVA and mean retinal sensitivity compared with the subgroup with foveal hypo-FAF. The integrity of both the photoreceptor inner/outer segment junction and the photoreceptor outer segment/retinal pigmented epithelium junction was significantly correlated with a preserved BCVA and a foveal hyper-FAF pattern on NIR-FAF. The findings suggested that NIR-FAF patterns correlate with morpho-functional outcomes in eyes affected by Stargardt disease.

Biochemical Features

Radu et al. (2003) demonstrated that treatment with isotretinoin (Accutane), an agent used in the treatment of acne, slows the accumulation of lipofuscin pigments in the eyes of Abcr-null mice. The results corroborated a proposed mechanism of biogenesis of N-retinylidene-N-retinylethanolamine (A2E), the main lipofuscin pigment that accumulates in cells of the RPE in patients with STGD. Radu et al. (2003) suggested that treatment with isotretinoin may inhibit lipofuscin accumulation in patients with STGD and thus delay the onset of visual loss.

Commenting on the work of Radu et al. (2003), Sparrow (2003) provided an explanation for the fact that the RPE cells underlying the macula have the highest accumulation of lipofuscin and that STGD primarily involves the center of the field of vision. He suggested that it is not a coincidence that the macula of the retina also has the highest concentration of 11-cis-retinal-containing visual pigment, a feature reflecting, in part, the packing density of cone and rod photoreceptor cells. The heightened capacity for photon absorption conferred by the density of visual pigment in the macula translates into a higher probability that all-trans-retinal will be available for A2E formation. Sparrow (2003) noted that the well-known cause of birth defects by orally administered isotretinoin would be problematic for female patients with STGD who might be treated with this agent.

Shroyer et al. (2001) analyzed DNA from 8 patients with clinically confirmed chloroquine or hydroxychloroquine retinopathy. Two patients had heterozygous ABCA4 mutations previously associated with Stargardt disease. None of the 80 controls had these missense mutations. Three other patients had other missense polymorphisms. The authors concluded that some individuals with ABCA4 mutations may be predisposed to develop retinal toxicity when exposed to chloroquine/ hydroxychloroquine, and they urged further study of a larger cohort of patients with this retinopathy.

Pathogenesis

Allikmets et al. (1997) commented that the accumulation in the retinal pigment epithelium (RPE) of a lipofuscin-like substance in STGD suggests that the site of ABCR-mediated transport may be on the apical face of the photoreceptor cell and that this transport may affect exchange between the RPE and the photoreceptors. The RPE participates in the continual renewal of visual pigments and of photoreceptor outer segments. The best-studied molecules that cycle between photoreceptors and the RPE are the retinoids. Allikmets et al. (1997) commented that if ABCR is involved in either export or import of retinoids, mutations in ABCR should lead to an accumulation of retinoids or their derivatives in the outer segment or the RPE, respectively. Histopathologic studies of the eyes in Stargardt disease, and its somewhat milder variant fundus flavimaculatus (FFM), show massive accumulation of lysosomal material similar to lipofuscin within RPE cells. Birnbach et al. (1994) additionally emphasized abnormal photoreceptor morphology and abnormal accumulation of lipofuscin in photoreceptor segments.

Cideciyan et al. (2004) studied surrogate measures of retinoid cycle kinetics, lipofuscin accumulation, and rod and cone photoreceptor and RPE loss in STGD1 and CORD3 (604116) patients with ABCA4 mutations and a wide spectrum of disease severity. There were different extents of photoreceptor/RPE loss and lipofuscin accumulation in different regions of the retina. Slowing of retinoid cycle kinetics was not present in all patients; when present, it was not homogeneous across the retina; and the extent of slowing correlated well with the degree of degeneration. The orderly relationship between these phenotypic features permitted the development of a model of disease sequence in retinal degeneration due to ABCA4 mutation, which predicted lipofuscin accumulation as a key early component of disease expression with abnormal slowing of the rod and cone retinoid cycle occurring at later stages of the disease sequence.

Population Genetics

Stargardt disease is the most common hereditary recessive macular dystrophy, with an estimated incidence of 1 in 10,000 (Blacharski, 1988).

Mapping

By genetic linkage analysis, using (CA)n microsatellite markers of known chromosomal location (Weissenbach et al., 1992) in 8 families, Kaplan et al. (1993) assigned the STGD locus to 1p21-p13. The combined maximum lod score was 6.88 at theta = 0.02 for the D1S236 locus.

From linkage studies, Gerber et al. (1995) concluded that fundus flavimaculatus with macular dystrophy and Stargardt disease are probably allelic disorders despite differences in age at onset, clinical course, and severity. In 4 families with late-onset FFM with macular dystrophy they found linkage to 1p21-p13, in the genetic interval defined by microsatellite loci D1S435 and D1S415; maximum lod score = 4.79 at theta = 0.0 for D1S435. They considered 1p13 to be the likely location of the gene that is mutant in these allelic disorders. The age at onset ranged from 17 to 60 years in adult patients.

By combined linkage analysis of 47 families with autosomal recessive STGD and/or FFM, Anderson et al. (1995) found significant linkage to marker D1S188 on chromosome 1p (maximum lod score of 32.7). Analysis of recombinants localized the disease locus to a 4-cM interval between D1S435 and D1S236. The findings indicated genetic homogeneity for STGD and FFM.

Hoyng et al. (1996) carried out linkage analysis in 7 families with recessive Stargardt disease and confirmed the location of a major recessive STGD gene on chromosome 1p22-p21. The maximum 2-point lod score for all families combined was 5.35 at theta = 0.04 for the marker D1S188 and the disease locus. Hoyng et al. (1996) genotyped 9 markers in 7 families to construct haplotypes, and this enabled them to reduce the STGD critical region to 2 cM, flanked by D1S406 and D1S236. In 1 family they encountered an affected female who, on the basis of haplotype analysis, carried only 1 disease allele. Hoyng et al. (1996) proposed several possible explanations for this finding, including the occurrence of genetic heterogeneity and the possibility that dominant and recessive mutations may be observed at 1 gene locus. They reported apparent nonpenetrance in a 45-year-old male.

Molecular Genetics

Allikmets et al. (1997) performed mutation analysis of the ABCA4 gene in STGD families and identified a total of 19 different mutations, including homozygous mutations in 2 families with consanguineous parentage (see, e.g., 601691.0002).

Shroyer et al. (1999) analyzed the ABCA4 gene in a 3-generation family manifesting both Stargardt disease and age-related macular degeneration (ARMD; see 153800), and identified heterozygosity for a missense mutation (P1380L; 601691.0026) in the paternal grandmother with ARMD, whereas the proband and his 2 paternal cousins with Stargardt disease were compound heterozygous for the P1380L mutation and another missense mutation in the ABCA4 gene (601691.0036 and 601691.0037, respectively). Shroyer et al. (1999) suggested that carrier relatives of STGD patients may have an increased risk of developing ARMD.

Single-copy variants of the ABCA4 gene have been shown to confer enhanced susceptibility to ARMD. By mutation analysis in a cohort of families that manifested both STGD and ARMD, Shroyer et al. (2001) found that ARMD-affected relatives of STGD patients are more likely to be carriers of pathogenic STGD alleles than predicted based on chance alone. Shroyer et al. (2001) used an in vitro biochemical assay to test for protein expression and ATP-binding defects, and found that mutations associated with ARMD have a range of assayable defects ranging from no detectable defect to apparent null alleles. Of the 21 missense ABCA4 mutations reported in patients with ARMD, 16 (76%) showed abnormalities in protein expression, ATP-binding, or ATPase activity. They inferred that carrier relatives of STGD patients are predisposed to develop ARMD.

Bernstein et al. (2002) examined 19 of 33 sibs from 15 Stargardt families who carried their respective proband's variant ABCA4 allele. Some families exhibited concordance of ABCA4 alleles with the macular degeneration phenotype, but others did not. Exudative ARMD was uncommon among both probands and sibs.

Fingert et al. (2006) reported a case of Stargardt disease in a patient homozygous for a mutation in the ABCA4 gene (601691.0026) as a result of uniparental isodisomy of chromosome 1. The patient's father was heterozygous for the mutation.

In a patient with juvenile macular degeneration in whom mutation in the ABCA4 gene was excluded (Briggs et al., 2001), Nishiguchi et al. (2005) identified mutations in the CNGB3 gene (605080.0002 and 605080.0006). Stargardt disease had not previously been associated with mutations in the cone channel subunits.

Pal Singh et al. (2006) identified homozygous null ABCA4 mutations (601691.0028-601691.0029) causing autosomal recessive nonsyndromic retinal dystrophy in 2 Indian families. Affected individuals in both families had early-onset visual loss, diminished rod and cone electroretinographic responses, and widespread atrophy of the retinal pigment epithelium.

In a retrospective study of 42 patients diagnosed with STGD in childhood, Fujinami et al. (2015) detected 2 or more disease-causing ABCA4 variants in 80% of the children. They also found a higher proportion of definitely or possibly deleterious variants in these children compared with 64 adult-onset ABCA4-positive STGD patients.

Genotype/Phenotype Correlations

Rozet et al. (1998) reported a genotype/phenotype correlation in ABCA4 gene mutations. They found that nonsense mutations truncating the ABCA4 protein consistently led to Stargardt disease, whereas all mutations they identified in the ABCA4 gene in fundus flavimaculatus were missense mutations affecting uncharged amino acids.

In a family segregating retinitis pigmentosa-19 (RP19; 601718) and STGD1 in 2 first cousins, Rozet et al. (1999) found that heterozygosity for a splice acceptor site mutation in the ABCA4 gene (601691.0017) resulted in STGD1, whereas hemizygosity for this mutation resulted in RP19. In the patient with RP19, a partial deletion of the maternal ABCA4 gene was presumed to be the source of a null allele, although this was not conclusively proven.

Duncker et al. (2015) used ABCR600 microarray analysis, next-generation sequencing, or both, to screen 37 patients with what the authors called 'bull's eye maculopathy' for mutations in the ABCA4 gene. ABCA4 mutations were identified in 22 patients, who tended to be younger than those without ABCA4 mutations. Whereas phenotypic differences were not obvious on the basis of either qualitative fundus autofluorescence (AF) or SD-OCT, with quantitative AF (qAF), the 2 groups of patients were clearly distinguishable. In the ABCA4-positive group, 37 of 41 eyes (19 of 22 patients) had qAF-8 of more than the 95% confidence interval for age, whereas in the ABCA4-negative group, 22 of 26 eyes (13 of 15 patients) had qAF-8 within the normal range.

History

Stargardt disease has sometimes been called central retinitis pigmentosa or retinitis pigmentosa with macular involvement. However, ordinary retinitis pigmentosa does not affect the macula.