Best Disease and Bestrophinopathies

Best vitelliform macular dystrophy (BVMD), BEST1-associated adult-onset vitelliform macular dystrophy, autosomal recessive bestrophinopathy (ARB) and autosomal dominant vitreoretinochoroidopathy (ADVIRC); adult-onset foveomacular vitelliform dystrophy due to PRPH2, IMPG1 and IMPG2 as a subsection and differential diagnosis.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

The abbreviations in the table are explained in the text: RPE, retinal pigment epithelium; EOG, electro-oculogram; ERG, electroretinogram; SD-OCT, spectral-domain optical coherence tomography; CNV, copy number variation (in genetic testing) or choroidal neovascularisation (in the fundus); anti-VEGF, vascular endothelial growth factor inhibitors; VA, visual acuity.

1. Introduction

1.1. From Best's description to identification of the gene

In 1905, the German ophthalmologist Friedrich Best described a family with a hereditary macular disorder, published in the Zeitschrift für Augenheilkunde under the title «Über eine hereditäre Maculaaffektion» [33]. The designation «Best disease» became established from that observation, and the yellow, round, well-demarcated macular lesion, compared to the yolk of a fried egg, gave rise to the term «vitelliform» (from the Latin vitellus, yolk). For much of the twentieth century the entity was defined clinically and, from the 1960s and 1970s onwards, electrophysiologically: the marked reduction of the light rise on the electro-oculogram (EOG), with a normal full-field electroretinogram (ERG), became the functional marker of the disease and allowed carriers with a normal fundus to be recognised [1, 25]. A population survey carried out in the county of Västerbotten, in northern Sweden, and published in 1974, documented a high concentration of hereditary macular degeneration, with 95 affected individuals, 75 of whom belonged to a single family, which suggests a founder effect [20, 34].

The molecular era began with the mapping of the vitelliform macular dystrophy locus to 11q12–q13.1. In 1998, Stöhr and colleagues constructed a transcript map with 19 genes in an interval of about 1.4 Mb [35]. In the same year, two groups independently identified the responsible gene. Petrukhin and colleagues, by analysis of recombination breakpoints in families that included the large Swedish pedigree designated «pedigree S1», found five independent disease-specific variants in a gene expressed in the retina, which they named VMD2 [36]. Marquardt and colleagues narrowed the critical interval to 980 kb between marker D11S4076 and the uteroglobin gene and detected heterozygous variants in the 12 probands of the large multigenerational families studied: ten missense substitutions at amino acids conserved from Caenorhabditis elegans to humans (published designations V9M, A10T, W24C, R25Q, R218Q, Y227N, Y227C, V235M, P297A and F305S; in NM_004183.4, according to ClinVar, c.25G>A p.(Val9Met), c.28G>A p.(Ala10Thr), p.(Trp24Cys), c.74G>A p.(Arg25Gln), c.679T>A p.(Tyr227Asn), c.680A>G p.(Tyr227Cys), c.703G>A p.(Val235Met), c.889C>G p.(Pro297Ala) and c.914T>C p.(Phe305Ser); R218Q remains as a published designation, with no corresponding ClinVar record) and a single-codon deletion, ΔI295 (NM_004183.4:c.884_886del p.(Ile295del)), in two families, with transcript expression restricted to the retinal pigment epithelium (RPE) by Northern blot [37]. The gene was later renamed BEST1 and the protein bestrophin-1.

Subsequent milestones broadened the spectrum. In 2000, Krämer and colleagues found 23 distinct variants in 41 unrelated patients with Best disease, four variants in 32 patients with adult-onset vitelliform macular dystrophy (two of them also present in patients with Best disease, which established an aetiological overlap) and no variants in 200 patients with age-related macular degeneration (AMD) [8]. In 2002, Sun and colleagues showed that human, Drosophila and C. elegans bestrophins form oligomeric chloride channels and that the human one is sensitive to intracellular calcium, framing Best disease as a channelopathy [38]. In 2004, Yardley and colleagues associated autosomal dominant vitreoretinochoroidopathy (ADVIRC) with nanophthalmos with BEST1 variants that alter splicing [4]. In 2008, Burgess and colleagues described autosomal recessive bestrophinopathy (ARB) and proposed it as the null BEST1 phenotype in humans [3]. In 2009, Davidson and colleagues attributed a form of retinitis pigmentosa, RP50, to BEST1 [39]. In 2014 the first crystal structures of bestrophins, chicken and bacterial, were solved, showing a pentameric channel [40, 41]. In September 2025 the first gene therapy clinical trial specific to bestrophinopathies, BIRD-1, began [28].

1.2. Nomenclature and reclassifications

The term «bestrophinopathies», used by GeneReviews, groups the diseases caused by pathogenic variants in BEST1 [1]. The former genetic designation VMD2 (vitelliform macular dystrophy 2) persists in the titles of articles predating about 2008 and in the name of OMIM entry 153700 («Macular dystrophy, vitelliform, 2») [1]. The numbers VMD3, VMD4 and VMD5 designate the vitelliform macular dystrophies associated with PRPH2, IMPG1 and IMPG2, respectively, and not variants of Best disease [6, 10, 11]. The name «autosomal recessive Best vitelliform macular dystrophy», used in publications from 2011–2012, refers to what is now called ARB, and the entity «retinitis pigmentosa 50» is disputed by GeneReviews itself [1, 2]. Adult-onset foveomacular vitelliform dystrophy appears with the abbreviations AOFVD (Orphanet) and AFVD (clinical literature); AFVD here designates the clinical entity of any cause and AVMD the BEST1-associated form described in GeneReviews [1, 5].

1.3. Definition of the entities in scope

Best vitelliform macular dystrophy (BVMD) is an autosomal dominant macular dystrophy, usually with onset between 3 and 15 years of age, characterised by a unilateral or bilateral macular vitelliform lesion that passes through a sequence of stages up to atrophy or a fibrotic scar, with an abnormal EOG and a normal full-field ERG [1]. BEST1-associated adult-onset vitelliform macular dystrophy (AVMD) shares the appearance of the lesion, but manifests between 30 and 50 years of age, has smaller lesions and a normal or mildly reduced EOG [1]. ARB is the autosomal recessive form, with multifocal subretinal deposits, intraretinal and subretinal fluid, hyperopia and a predisposition to angle-closure glaucoma, a frequently reduced ERG and a severely abnormal EOG [1, 3]. ADVIRC is an autosomal dominant disease with involvement of the whole globe: a circumferential band of hyperpigmentation in the peripheral retina, with a sharp demarcation line, white preretinal opacities, fibrillary condensation of the vitreous and association with nanophthalmos, microcornea, hyperopia and a narrow angle [1, 23]. MRCS syndrome (microcornea, rod–cone dystrophy, cataract and posterior staphyloma; ORPHA:263347) is considered by Orphanet to be an allelic form linked to BEST1, sharing OMIM entry 193220 with ADVIRC [42]. Adult-onset foveomacular vitelliform dystrophy is a clinical phenotype with genetic heterogeneity, in which BEST1 explains a minority of cases and PRPH2, IMPG1 and IMPG2 another part, with most remaining without an identified cause [1, 24, 43].

Retinitis pigmentosa and inherited retinal dystrophies, Stargardt disease and ABCA4-associated retinopathies, and cone and cone–rod dystrophies are the differential diagnoses with which bestrophinopathies are most often confused, and MELAS and the m.3243A>G variant spectrum can present with vitelliform lesions in adults.

2. Molecular Genetics and Pathophysiology

2.1. The BEST1 gene and bestrophin-1

The BEST1 gene lies at 11q12.3 (HGNC:12703; NCBI Gene 7439; Ensembl ENSG00000167995) and the MANE Select (Matched Annotation from NCBI and EMBL-EBI) reference transcript is NM_004183.4, which encodes the protein NP_004174.1, with 585 amino acids [7, 23, 44]. GeneReviews still uses the earlier version NM_004183.3 [1]. The protein has no signal peptide, so the historical amino acid numbering coincides with the current Human Genome Variation Society (HGVS) numbering; this coincidence was confirmed in VariantValidator for p.(Tyr227Asn) and p.(Ile295del), two of the variants described by Marquardt in 1998 [7, 37].

Bestrophin-1 is expressed mainly in the RPE, where it is located in the basolateral plasma membrane [1]. It is an anion channel activated by intracellular calcium, permeable to chloride and bicarbonate; at rest it adopts a partially open conformation [44]. Beyond its channel function, bestrophin-1 regulates L-type voltage-dependent calcium channels (CaV1) through SH3-binding domains, and its dephosphorylation by the phosphatase PP2A suggests participation in a transduction pathway related to modulation of the EOG light peak [1]. The structure of chicken bestrophin-1, solved at 2.85 Å, showed a pentamer with a large cytosolic region where calcium binds (the «Ca²⁺ clasp»), a single pore about 95 Å long and at least 15 anion-binding sites, and a hydrophobic «neck» that acts as a gate; pathogenic variants cluster in the gating apparatus [40]. The structure of the bacterial homologue KpBest, also pentameric and with the pore closed by three hydrophobic rings, served as the basis for a homology model of the human protein that locates the pathogenic variants [41].

2.2. From channel to phenotype

The tissue mechanism common to the bestrophinopathies is dysfunction of RPE ion transport. Channel failure leads to accumulation of fluid and of material derived from photoreceptor outer segments between the RPE and the neurosensory retina, with lipofuscin accumulation within and beneath the RPE, which forms the vitelliform material [1, 36, 44]. In the natural canine model of bestrophinopathy (designated cmr or cBest), Guziewicz and colleagues showed that the visible clinical lesion is only the most apparent part of a retina-wide alteration: there is a defect in the interaction between the apical RPE microvilli and the cone outer segments, with insoluble cone-associated interphotoreceptor matrix, and a diffuse microdetachment between the RPE and the photoreceptors, which contracts with dark adaptation and expands with moderate light [45]. In patients with biallelic variants, the same authors observed large regions with lamination defects, severe loss of photoreceptor sensitivity and slowing of the retinoid cycle [45]. The measurable functional consequence is a reduced EOG light rise, which reflects loss of depolarisation of the basolateral RPE membrane [25, 44].

2.3. Molecular mechanisms by entity

GeneReviews describes the mechanism of the bestrophinopathies as loss of function: disease-associated missense variants reduce or abolish membrane current [1]. Subsequent research distinguishes three mechanisms, with direct consequences for the choice of gene therapy [23]. Simple loss of function characterises most ARB variants; Burgess and colleagues showed that two ARB missense variants severely reduce channel activity without a dominant-negative effect when co-expressed with the wild-type protein, and that heterozygotes have no clinical or electrophysiological changes, which led to ARB being proposed as the null phenotype [3]. In BVMD, Sun and colleagues observed that four mutants co-expressed with the wild-type protein inhibited it dominantly [38], and the incorporation of mutant subunits into a pentameric channel explains how a dominant-negative effect renders the channel non-functional or unstable; allelic expression imbalance at the BEST1 locus favours this effect [23]. Zhao and colleagues showed that some so-called dominant variants behave as dominant-negative and that loss-of-function and gain-of-function variants require different mutant:wild-type ratios [46]. ADVIRC, finally, is associated with variants that cause exon skipping and produce shortened isoforms with an internal deletion; according to the review by Haldrup and colleagues, RPE cells derived from induced pluripotent stem cells (iPSC) of patients with ADVIRC have increased anion transport, which amounts to gain of function [4, 23, 47]. In patient-derived RPE cells with six dominant BVMD variants, all showed loss of function, to different degrees and of different types [48]. From the cell-biology standpoint, proteins with recessive variants are recognised in the endoplasmic reticulum and degraded more rapidly, whereas dominant variants escape this quality control and are degraded in the Golgi apparatus (indirect attribution, according to Beryozkin and colleagues) [20].

2.4. Variant spectrum

The Leiden Open Variation Database (LOVD) had registered, in January 2024, 485 BEST1 variants, of which 366 (75%) were classified as pathogenic or likely pathogenic and 44 (9%) had conflicting interpretations that include that class [23]. The review by Beryozkin and colleagues, which integrated LOVD3, the Human Gene Mutation Database (HGMD) and the literature up to January 2023, counted 351 different missense variants (2141 alleles; 251 dominant and 100 recessive), 17 nonsense (74 alleles), 16 splicing (63 alleles), 27 frameshift (87 alleles) and 16 in-frame deletions or insertions (85 alleles) [20]. Nonsense, splicing and frameshift variants are associated with recessive disease; in-frame deletions and insertions are generally recessive, except when they affect amino acids 295–304, a region where they tend to be dominant [20]. Missense variants cluster in the N-terminal half of the protein. The least tolerant regions are the transmembrane regions (amino acids 31–50, 71–94, 235–257 and 269–287) and the intracellular regions (1–30, 95–234 and the first approximately 35 amino acids of segment 288–585); the extracellular regions (51–70 and 258–268) have only five missense variants described [20]. In domains IC1, TM2 and IC3, more than 80% of variants cause BVMD; domain IC2 concentrates more recessive variants (n=71), and in TM1, EC1 and TM4 more than 75% are recessive. About 23% of published missense variants are recessive and about 10% have been described with both modes of inheritance [20]. Worldwide, 90.1% of alleles are missense and 32.3% of published alleles are recessive [20].

The following table brings together the variants cited throughout this article, normalised in VariantValidator against the MANE Select transcript and the GRCh38 genome [7]. Where the source published a nucleotide notation that does not match the current reference sequence, the table records the discrepancy and does not reproduce it as valid nomenclature.

2.5. Genotype–phenotype correlations

For most BEST1 variants no genotype–phenotype correlation has been demonstrated, and the same variant produces different pictures within the same family [1]. The exceptions described are few. The variant p.(Val89Ala) was associated with late visual loss, at 40–50 years, in one family [49]; p.(Tyr227Asn) with small, late vitelliform lesions in a family also studied by clinicopathological correlation [50]; and p.(Ala243Val) with mild or late disease, with a distinct autofluorescence pattern [51–53]. In Norway, the most frequent genotypes, c.89A>G p.(Lys30Arg) and c.403G>A p.(Gly135Ser), were associated with BVMD of later onset than the remaining genotypes [55]. In ARB, the variant p.(Thr363Pro) was found only in disease limited to the macula in a French cohort of 34 patients [60]. The relationship between variant type and mode of inheritance is the most robust correlation: protein-truncating variants and most in-frame deletions cause recessive disease, missense variants in specific intracellular and transmembrane domains cause mainly dominant disease, and variants that alter splicing of exon 6 cause ADVIRC [9, 20]. Burgess and colleagues showed that variants c.704T>C and c.707 alter an exonic splicing enhancer and increase binding of the factor ASF/SF2, whereas variant c.703G>C, associated with Best disease, does not alter splicing [9]. Yardley and colleagues, using minigene assays, found that the three changes identified in five families with ADVIRC simultaneously produced the missense substitution and exon skipping, generating three isoforms (wild-type, missense and in-frame deletion) [4].

2.6. Penetrance and expressivity

The penetrance of BVMD is high but incomplete, above 70%, particularly when an abnormal EOG is accepted as evidence of expression [1]. Expressivity varies widely between and within families: heterozygotes may have a normal fundus with an abnormal EOG, unilateral lesions, multifocal lesions or atypical peripheral forms, such as the circinate mid-peripheral vitelliform dystrophy with macular sparing described in the father of a patient with Best disease [1, 61]. For this reason, the absence of an apparent family history does not exclude dominant inheritance, and examination of the parents' fundi is unreliable for excluding carrier status [1, 62]. In ARB, heterozygotes generally have no clinical disease; several studies have nonetheless described a slight reduction of the multifocal ERG (mfERG) or subtle macular symptoms in some carriers [1, 63–65].

2.7. Allelic disorders and controversial entities

BEST1-associated retinitis pigmentosa (RP50; OMIM 613194) was proposed by Davidson and colleagues on the basis of four missense variants (three novel) in patients with rod–cone dystrophy; p.(Leu140Val) and p.(Ile205Thr) reduced chloride currents and p.(Leu140Val) and p.(Asp228Asn) displaced the protein from the basolateral membrane to the cytoplasm [39]. GeneReviews considers the diagnosis of retinitis pigmentosa disputed and cites the hypothesis of a multigenic aetiology, in which the BEST1 variant would be an incidental finding or a modifier [1, 66]. The inheritance of this entity is not agreed: OMIM records it as autosomal recessive, whereas in GenCC two laboratories submit it as autosomal dominant, with «Strong» evidence in one case and «Limited» in the other [44]. The GeneReviews entry on nonsyndromic retinitis pigmentosa includes BEST1 among the rare genes, accounting for 1% or fewer of cases [17]. The ClinGen consortium classified the validity of the relationship between BEST1 and dominant retinopathy as «Definitive», with no actionability assertion [44].

MRCS syndrome has been described in families with BEST1 variants; in 2006, Michaelides and colleagues described a 12-year-old boy with MRCS without a variant in VMD2, demonstrating genetic heterogeneity, and noted that all previous cases had variants in this gene [67]. Orphanet counts seven cases or families, with onset in childhood and dominant inheritance [42]. The association of BEST1 with nanophthalmos derives from ADVIRC and indicates a role of the RPE and of bestrophin-1 in growth and patterning of the globe [4]; hyperopia and short axial length are also frequent in BVMD and ARB [22, 23]. A 2026 series of eight eyes of patients with ADVIRC found, contrary to the classical association, a median spherical refraction of −0.63 D and a median axial length of 24.7 mm, and the accompanying systematic review indicates that myopia is more prevalent than hyperopia in published ADVIRC [47].

2.8. Genes of adult-onset foveomacular vitelliform dystrophy

The PRPH2 gene (6p21.1; formerly RDS) encodes peripherin-2, a structural protein of the outer-segment disc rim. Its variants cause a spectrum that includes pattern macular dystrophy (MIM 169150), central areolar choroidal dystrophy (MIM 613105), vitelliform macular dystrophy 3 (MIM 608161), retinitis punctata albescens, retinitis pigmentosa 7 and the digenic PRPH2/ROM1 form, with marked intrafamilial variability; in the family described by Weleber and colleagues, the same deletion of codon 153 or 154 produced retinitis pigmentosa, pattern dystrophy and fundus flavimaculatus [10, 68, 69]. Felbor and colleagues found PRPH2 variants in 18% of 28 unrelated patients with adult-onset vitelliform macular dystrophy [43].

The IMPG1 gene (6q14.1) encodes an interphotoreceptor matrix proteoglycan (SPACR). In 2013, Manes and colleagues described the variant c.713T>G p.(Leu238Arg) in three families with dominant inheritance and, on screening 144 probands, a homozygous recessive form c.807+1G>T and a compound heterozygosity c.461T>C p.(Leu154Pro) with c.1519C>T p.(Arg507Ter); the Arden ratio was normal or moderately reduced in most [12]. ClinGen classifies the relationship of IMPG1 with dominant retinopathy and with recessive retinopathy as «Definitive» (2023), and the gene also causes retinitis pigmentosa 91 (MIM 153870) with early macular involvement [6, 70].

The IMPG2 gene (3q12.3) follows a dosage rule: biallelic variants cause retinitis pigmentosa 56 (MIM 613581) with early macular atrophy, and monoallelic variants cause vitelliform macular dystrophy 5 (MIM 616152) or pattern dystrophy. In a British series, six biallelic patients had a mean onset at 18.4 years and a mean visual acuity of 1.90 and 1.82 logMAR at 68.7 years, whereas five monoallelic patients (four with a vitelliform lesion and one with a pattern lesion) had a mean onset at 54.2 years and a mean visual acuity of 0.15 and −0.01 logMAR; the mechanism proposed for the monoallelic form is haploinsufficiency [11, 71]. Brandl and colleagues, in 106 patients negative for BEST1 and PRPH2, described three IMPG2 nonsense variants, p.(Glu226Ter), p.(Ser522Ter) and p.(Gln856Ter), and five missense variants, p.(Ala243Pro), p.(Gly1008Asp), p.(Phe1016Ser), p.(Tyr1042Cys) and p.(Cys1077Phe); asymptomatic carriers of p.(Ala243Pro) and of p.(Cys1077Phe) had subtle foveal irregularities [13]. The normalised reference variants are PRPH2 NM_000322.5:c.554T>C p.(Leu185Pro) (always present in digenic retinitis pigmentosa with ROM1), IMPG1 NM_001563.4:c.713T>G p.(Leu238Arg) and IMPG2 NM_016247.4:c.3230G>T p.(Cys1077Phe) [7, 10, 24].

3. Clinical Spectrum and Manifestations by System

The bestrophinopathies are exclusively ocular diseases. The sources consulted describe no systemic manifestations associated with BEST1, although bestrophin-1 has homologues expressed in other tissues [1, 44]. The spectrum is organised by entity, with differences in age at onset, retinal topography and anterior segment involvement that determine prognosis and surveillance.

3.1. Best vitelliform macular dystrophy: natural history by stage

BVMD is a slowly progressive macular dystrophy, usually with onset between 3 and 15 years of age [1]. The lesion may be unilateral or bilateral, single or multiple, macular or eccentric; multiple and extramacular lesions occur in at least 25% of patients [1]. Progression usually follows the sequence previtelliform, vitelliform, pseudohypopyon, vitelliruptive and atrophic or cicatricial, but not all patients pass through every stage and there is inter- and intrafamilial variability [1, 23]. The Gass classification, reproduced in GeneReviews, associates each stage with a fundus appearance and an angiographic pattern [1].

Sources: stages and angiography according to GeneReviews [1]; visual function according to the review by Haldrup and colleagues [23].

Visual acuity can go from 20/20 to less than 20/200 over a lifetime [1]. The review by Haldrup and colleagues states, citing earlier literature, that 75% of patients under 40 years of age retain a visual acuity of 6/12 or better in at least one eye; this is an indirect attribution and not a cohort read directly [23]. Rapid visual loss is associated mainly with choroidal neovascularisation (CNV) with haemorrhage [23]. Hyperopia, astigmatism and anterior segment changes have also been described in BVMD, possibly owing to the role of bestrophin-1 in ocular growth [23]. The 2021 French PNDS associates the hyperopia of Best disease with a risk of glaucoma in adults and describes a simplified sequence in three phases (vitelline, fragmentation and resorption), which may coexist in different lesions of the same eye; reduced visual acuity is noticed only when the vitelline material fragments, and acuity may improve after resorption, which makes Best disease the only genetic maculopathy in which the PNDS describes visual improvement during its course [21].

Recent cohorts add quantitative natural-history data. In a Chinese cohort of 134 patients from 115 families with BEST1 variants (60 with BVMD, from 48 families, and 74 with ARB, from 67 families), 58 of the patients with BVMD had typical vitelliform or atrophic changes and two had a pattern-dystrophy-like appearance or irregular RPE changes; CNV was present in 35.0% of eyes with BVMD, eyes with CNV had worse visual acuity and the decline in visual acuity was slow, particularly in the absence of CNV and angle-closure glaucoma [72]. In an Italian mesopic microperimetry study (MAIA, 68-point grid over the central 10°) with 23 eyes of 13 patients with BVMD, median age 29.4 years and a mean of three examinations over a median of 4.4 years, the median macular sensitivity at baseline was 21.4 dB, with 44% normal points, 54% relative scotomas and 2% deep scotomas; there was no significant overall decline (−0.16 dB/year; p=0.21), but initially normal points lost 0.36 dB/year (p=0.006), with greater loss in the central 5° (−0.56 dB/year; p=0.028), and on average one new deep-scotoma point appeared every five years (0.22 points/year; p=0.015) [73]. In a series of 15 patients with confirmed BVMD (10 children and 5 adults), the children had less advanced stages and better visual acuity (0.13 ± 0.17 logMAR versus 0.91 ± 0.52); the mfERG showed macular dysfunction in all, even in children with 20/20, with a 51% reduction of the P1 wave in ring 1 in both children and adults, and correlated more with the degree of vitelliform liquefaction on optical coherence tomography (OCT) than with visual acuity; the full-field ERG was normal in the children and reduced or delayed in some adults [74].

The disease is not confined to the posterior pole. In an ultra-widefield imaging study of 72 eyes of 36 patients (58 eyes of 29 patients with BVMD and 14 eyes of 7 patients with ARB) compared with 78 eyes of 39 controls, patients were more likely to have peripheral changes; the most frequent were areas of «white without pressure» and «dark without pressure», particularly in ARB (86% of eyes), with attenuation of the ellipsoid zone in the dark areas without vitreous traction, and almost half of the eyes had vascular anomalies (intraretinal microvascular abnormalities and tortuosity more in BVMD; focal narrowing more in ARB), distributed across all Gass stages [75]. These data agree with the diffuse microdetachment observed in the canine model [45].

The histopathology of a 72-year-old man with molecularly confirmed Best disease, presented in GeneReviews, showed in the right eye loss of photoreceptors, an attenuated outer plexiform layer, periodic acid–Schiff (PAS)-positive deposits and pigment granules in the fovea without underlying RPE, with perifoveal RPE hyperpigmentation; in the left eye there was a small disciform scar of thin fibrous tissue with PAS-positive material, without RPE [1].

3.2. Choroidal neovascularisation in BVMD

GeneReviews classifies CNV as infrequent in BVMD [1]. OCT angiography (OCTA) has changed this perception. At the University of Iowa, between September 2017 and October 2021, Han and colleagues examined by swept-source OCTA 53 eyes of 27 patients with molecularly confirmed BVMD (14 men and 13 women; mean age 45 years, median 49, range 8–79; mean visual acuity 0.38 logMAR) and found CNV in 27 eyes (50.9%), all type 1 (under the RPE); only 11 of these 27 eyes (40.7%) had a previous clinical diagnosis of CNV [56]. CNV was present in 17 of 27 eyes (63.0%) with a typical vitelliform lesion and subretinal fluid (Gass stage 2), in no eye with a solid vitelliform lesion, and in all eyes with a fibrotic subretinal «pillar» (8/8) or focal choroidal excavation (8/8) (p<0.01); mean subfoveal choroidal thickness was 287.5 ± 126.9 µm and greater in eyes with CNV (323.2 versus 250.5 µm; p=0.04) [56]. In the longitudinal subgroup of seven patients, with a mean of 4.1 examinations over 21.3 months (12–37), CNV was detected in 10 of 14 eyes (71.4%) [56]. In a prospective study by Parodi and colleagues of 78 eyes with BVMD, macular neovascularisation was present in 50 eyes (64%) at baseline and in 51 (65%) at the end of a follow-up of 24.7 ± 9.7 months; in stages 2–3 there was neovascularisation in 4 of 30 eyes (13%), exudative, and in stages 4–5 in 46 of 48 eyes (96%), non-exudative, which remained stable with observation alone [76]. The authors therefore distinguish an exudative form, rare and early, with haemorrhage and fluid, and a non-exudative form, common and late [76]. The 2021 French PNDS gives a CNV frequency of about 15% in Best disease, without identifying the source series, and considers the complication very rare in Stargardt disease and constant at around 50 years of age in Sorsby dystrophy [21]. The difference between this value, the 35% of eyes with CNV in the Chinese cohort and the 51–65% of the OCTA series stems mainly from the detection method: the lower values correspond to clinically recognised CNV, and OCTA adds non-exudative neovascularisation, which in most eyes has no clinical expression [56, 72, 76].

Han and colleagues propose that CNV participates in the evolution of the lesion itself: neovascularisation beneath the vitelliform lesion would breach the RPE and allow fluid to drain into the choroid, giving rise to the vitelliruptive lesion; regression of CNV with anti-VEGF may reconstitute the fluid and the vitelliform lesion [56]. Acquired «torpedo»-like maculopathy associated with macular neovascularisation and paediatric CNV monitored by OCT and OCTA have been described in 2026 series [77, 78].

3.3. BEST1-associated adult-onset vitelliform macular dystrophy

AVMD belongs to the group of pattern dystrophies. Symptoms usually appear in the fifth decade, with typical onset between 30 and 50 years of age; some patients are asymptomatic and others have reduced visual acuity and metamorphopsia, with slow deterioration [1]. The yellow subfoveal deposits are about one third of a disc diameter in size, enlarge and then shrink until they leave an atrophic area; the subretinal lesion has heterogeneous hyperreflectivity and half of the eyes have pigment epithelial detachments [1]. Visual acuity ranges between 20/40 and 20/200 and CNV is rare [1]. Because of its age at onset, AVMD is frequently confused with AMD [23]. A 2026 case describes BVMD with the appearance of pachyvitelliform maculopathy in an older person [79].

3.4. Autosomal recessive bestrophinopathy

ARB typically begins in the first decade but may manifest up to the fifth, and is more severe than BVMD [1]. The fundus shows multifocal yellow subretinal deposits in the macula and periphery, with intraretinal and subretinal fluid in the early stages and subretinal fibrosis in the late stages; visual acuity ranges from normal to less than 20/200, depending on macular involvement [1]. Of the six children from three families summarised in GeneReviews, four had multiple vitelliform lesions [1]. In the Dutch series of Boon and colleagues, with 10 patients from 7 families, onset ranged between 2 and 54 years; all were hyperopic, some had narrow angles with a predisposition to angle-closure glaucoma, the EOG was abnormal in all and the full-field ERG abnormal in 8 and normal in 2 [22].

The French cohort of Bianco and colleagues (REFERET reference centre, Quinze-Vingts, Paris) included 34 consecutive patients (68 eyes) with molecular confirmation, median age at baseline 32 years (interquartile range 18.3–46.9) [60]. On short-wavelength autofluorescence, 21% (7/34) had isolated macular lesions (grade 1), 44% (15/34) multifocal lesions or diffuse changes of the posterior pole with zonal sparing (grade 2) and 35% (12/34) panretinal changes (grade 3); marked ERG attenuation occurred only in grade 3 [60]. Primary angle closure was present in 29% (10/34), and the median age at which severe visual impairment by World Health Organization criteria was reached was 47 years in patients with angle closure and 68 years in the others (p=0.01), with no association with autofluorescence grade; there was no significant annual linear change in visual acuity over a median follow-up of 3 years [60]. Visual risk in ARB is thus dominated by angle closure. In the Chinese cohort, 67 of 74 patients with ARB had scattered vitelliform lesions, 5 had severe retinal detachment and CNV was present in 25.5% of eyes [72]. Cases published in 2025–2026 describe acute angle-closure glaucoma as the first manifestation, short eyes with a shallow anterior chamber, macular schisis-type changes and central retinal vein occlusion associated with angle closure [30, 59, 80, 81].

3.5. Autosomal dominant vitreoretinochoroidopathy

ADVIRC is characterised by circumscribed hyperpigmentation of the peripheral retina, with a sharp demarcation line in the mid-periphery between normal and abnormal retina; the band lies between the vortex veins and the ora serrata and is the hallmark of the disease [1, 23]. There are white punctate preretinal opacities in the hyperpigmented areas, arteriolar narrowing, frequent cystoid macular degeneration, preretinal neovascularisation, and cells and fibrillary condensation in the vitreous that may obscure vision; visual acuity remains good until late stages, when the whole retina is involved [1, 23]. Ocular developmental abnormalities are associated: nanophthalmos, microcornea, hyperopia, presenile cataract, a shallow anterior chamber with angle-closure glaucoma and optic nerve dysplasia [1, 9, 23]. The series by Ni and colleagues compared 8 eyes with ADVIRC, 10 with BVMD and 10 controls (median ages 51, 38 and 50 years; median visual acuity 0.20, 0.02 and 0.00 logMAR): the outer-segment layer was thinner in ADVIRC than in BVMD at all locations and the photoreceptor layer was thinner than in BVMD and in controls in the perifovea [47].

3.6. MRCS syndrome and the retinitis pigmentosa phenotype

MRCS syndrome combines microcornea, rod–cone dystrophy, cataract and posterior staphyloma, with onset in childhood [42, 67]. The retinitis pigmentosa-like phenotype associated with BEST1 is disputed (section 2.7) [1, 39].

3.7. Non-BEST1 adult-onset foveomacular vitelliform dystrophy

In the IMPG1 and IMPG2 forms from the French reference centre (13 patients from 4 families), mean onset was at 42 years and mean visual acuity 20/40; drusen-like lesions adjacent to the foveal lesion were present in 9 of 13 patients, the vitelliform material lay above the RPE at all stages, with preserved RPE reflectivity, and the EOG was normal or borderline in 9 cases [24]. Brandl and colleagues described as the common picture a bilateral central dome-shaped foveal accumulation with an intact RPE, with more severe symptoms in the IMPG1 forms, and two simplex cases with onset in early childhood [13]. The PRPH2 forms manifest as macular and pattern dystrophies in the fourth and fifth decades, with a slow decline in visual acuity [10].

3.8. Frequency of findings by BEST1 phenotype

Source: table 2 of GeneReviews [1]. The frequency of CNV detected by OCTA in BVMD is higher than the clinical classification suggests (section 3.2) [56, 76].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age

In children and adolescents, suspicion arises from a yellow, round, well-demarcated macular lesion, sometimes discovered on a routine examination with normal visual acuity, or from reduced vision, metamorphopsia or photophobia [1, 23]. Multiple, extramacular or unilateral vitelliform lesions do not rule out the diagnosis [1]. A child with vitelliform lesions should be tested for BEST1 even when the parents have a normal EOG and normal imaging, because this is the expected pattern in ARB [2]. A difference in visual acuity between the eyes in a child with a unilateral macular lesion raises the suspicion of associated amblyopia [1].

In a young adult with hyperopia, short eyes and a shallow anterior chamber, angle-closure glaucoma associated with unexplained retinal changes (macular schisis, subretinal fluid, multifocal yellow deposits) should prompt consideration of ARB and a request for genetic testing [29, 59]. Between 30 and 50 years of age, a small, bilateral subfoveal vitelliform lesion suggests AVMD or AFVD of another cause and is frequently labelled as AMD [1, 23]. A hyperpigmented peripheral band with sharp demarcation, nanophthalmos or microcornea suggests ADVIRC [1].

The French PNDS on genetic maculopathies summarises the clinical features that point to a hereditary maculopathy: reduced central visual acuity not correctable with lenses, as a rule bilateral and symmetrical, with the exception of the vitelliform dystrophies, which may be unilateral or asymmetrical; a central scotoma, which manifests as difficulty recognising faces; and colour vision changes proportional to the loss of acuity [21]. Inaugural or intense photophobia points rather to a cone dystrophy, and restriction of the peripheral visual field or night blindness to a mixed dystrophy (cone–rod or rod–cone); metamorphopsia is rare in hereditary maculopathies outside the atrophic stages and CNV, so its appearance suggests a neovascular complication [21]. The absence of a family history does not exclude a genetic origin, but, in an isolated case, the PNDS requires acquired causes to be ruled out: drugs (synthetic antimalarials, antiretrovirals, dietary supplements), recreational drugs (inhaled nitrites, or poppers), exposure to recreational lasers or to sunlight, diabetes, vascular occlusions and inflammatory or infectious diseases (toxoplasmosis, multifocal choroiditis, placoid epitheliopathy, cat-scratch disease) [21]. Extraocular signs point to syndromic forms that may begin with macular involvement: polydactyly, obesity, intellectual disability and renal abnormalities in Bardet–Biedl syndrome; developmental regression, behavioural changes and epilepsy in neuronal ceroid lipofuscinosis; gait disturbance, dysarthria and dysphagia in spinocerebellar ataxia type 7; and, in adults, maternally inherited deafness and diabetes with progressive multisystem involvement in maternally inherited diabetes and deafness (MIDD) [21].

4.2. Warning and emergency signs

A recent decrease in visual acuity or metamorphopsia in a patient with a known vitelliform lesion suggests exudative CNV or transition to the vitelliruptive stage and warrants prompt ophthalmological assessment with OCT and OCTA [1, 56]. Subretinal haemorrhage is a sign of active CNV [1]. Acute angle-closure glaucoma, which may be the first manifestation of ARB, is an ophthalmological emergency in a patient with ARB, ADVIRC or BVMD with short eyes [29, 59]. Severe retinal detachment was observed in 5 of 74 patients with ARB in a Chinese cohort [72].

4.3. Newborn screening and population screening

There is no newborn or population screening for the bestrophinopathies. Relatives at risk are identified by cascade testing and ophthalmological examination (section 8) [1].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic criteria

No published consensus diagnostic criteria exist for the bestrophinopathies [1]. GeneReviews establishes the diagnosis in a proband with suggestive clinical and electrophysiological findings and a heterozygous pathogenic or likely pathogenic variant in BEST1 (dominant forms) or biallelic variants (ARB) [1]. Pathogenic and likely pathogenic variants are clinically equivalent for this purpose, and a variant of uncertain significance (VUS) neither establishes nor excludes the diagnosis [1]. The family history is usually dominant; affected siblings or consanguinity suggest ARB; the absence of a family history excludes none of the forms [1].

Source: suggestive findings according to GeneReviews [1]; ARB according to Boon and colleagues and the ISCEV standard [22, 25].

The 2021 French PNDS lists clinical criteria for Best disease close to those of GeneReviews, with some practical refinements [21]. Discovery may be incidental, before any complaint; the patient is often hyperopic; the fundus shows round yellow deposits, single or multiple, macular or extramacular, unilateral or bilateral and sometimes at different stages in the same eye; SD-OCT shows retinal thickening with hyperreflective material until resorption, after which the retina may be thinned; autofluorescence shows strongly hyperautofluorescent material; the full-field ERG has normal amplitudes and implicit times, except in the recessive form; and the EOG, abnormal in most patients (reduced Arden ratio), is the test that, combined with testing for the familial variant, allows asymptomatic carriers to be recognised, although a normal EOG does not exclude the diagnosis [21]. In the recessive form, the PNDS highlights multiple deposits, cystic lesions, multiple serous detachments, extension of the deposits beyond the vascular arcades and an abnormal ERG, and considers it more severe than the dominant form [21].

5.2. Electro-oculogram

The EOG measures the standing potential of the eye during standardised saccadic movements, first in dark adaptation and then in light adaptation. The standard of the International Society for Clinical Electrophysiology of Vision (ISCEV), updated in 2017, specifies 15 minutes of dark adaptation followed by 15 minutes of light adaptation, with a photopic background luminance of 100 cd·m⁻² (acceptable 90–110), dilated pupils with recording of the diameter and at least 30 minutes of stable room lighting before the test, avoiding indirect ophthalmoscopy, autofluorescence and fundus photography during that period [25]. ISCEV recommends the term «LP:DT ratio» (light peak/dark trough), equivalent to the Arden ratio. Each laboratory should establish its own normative values (median and 5th–95th percentiles); the normal LP:DT ratio typically lies between 1.7 and 4.3 and the time to light peak between 7 and 12 minutes [25]. The report should include the LP:DT ratio, the dark trough amplitude in µV and the time to peak [25]. GeneReviews gives an Arden ratio above 1.8 as normal and, in BVMD, values below 1.5, almost always between 1.0 and 1.3 [1]. The ratio decreases with age from the fourth decade onwards, so the threshold is not absolute [1].

The ISCEV standard summarises the diagnostic role of the EOG in the bestrophinopathies: in Best disease, the full-field ERG is usually normal and the LP:DT ratio abnormal, and an abnormal EOG can distinguish Best disease from other dominant diseases with a similar fundus, including some cases of adult-onset vitelliform macular dystrophy and pattern dystrophy; in ARB and ADVIRC, the ERG is frequently abnormal, but the EOG is severely or disproportionately abnormal [25]. In most other retinal diseases, EOG changes are proportional to those of the rod ERG and have no diagnostic value of their own [25]. In vitelliform dystrophy due to IMPG1 or IMPG2, the EOG is normal or borderline, which points the diagnosis towards these genes [6, 24].

The limitations of the EOG are practical and biological. The EOG may occasionally be normal in molecularly confirmed BVMD, as in the Italian family with a de novo variant described by Testa and colleagues [1, 82]. Cross-talk between the eyes is about 15% with electrodes near the inner canthi and may reach about 40% if the electrodes touch; patients who cannot fixate, because of poor central vision or nystagmus, and young children may not cooperate with the saccadic movements [25]. The examination is time-consuming, which has prompted shortened protocols, with recording during the ERG or shorter adaptation [83]. The short-wavelength EOG (448 nm at 30 cd/m²) was reduced in four of four cases of genetically confirmed BVMD and considered more comfortable by patients, but this is a preliminary result that requires validation [84].

5.3. Electroretinography

The full-field ERG is normal in BVMD and AVMD and reduced, with diminished scotopic and photopic responses, in ARB; in practice, a reduced full-field ERG points to the recessive form [1, 44]. In the French ARB cohort, marked ERG attenuation occurred only in panretinal disease [60]. The American Academy of Ophthalmology (AAO) considers the full-field ERG unnecessary in Best disease, but states that, when the EOG is not available, the full-field ERG should be normal in Best disease [85]. The French PNDS considers the full-field ERG according to the ISCEV standard indispensable at any age in the presence of a maculopathy, to measure severity, estimate progression and separate maculopathies from cone dystrophies and mixed dystrophies, with a single exception: the typical picture of Best disease, with characteristic vitelliform material and autosomal dominant transmission [21]. The multifocal ERG and the foveal ERG show reduced central amplitudes, and the abnormal areas on mfERG coincide with the abnormal areas on OCT and fundus photography [1]. The mfERG is abnormal even in children with a visual acuity of 20/20 [74]. Colour vision shows, in a significant proportion of patients, anomalous discrimination mainly along the protan axis, a non-specific finding with no diagnostic value [1].

5.4. Multimodal imaging

SD-OCT is the central staging examination: in the previtelliform stages the architecture is normal or shows subtle outer retinal changes; in the intermediate stages there is separation and elevation of the outer retina–RPE complex, with dome-shaped material and subretinal fluid; in the atrophic stage there is thinning of the retina and RPE [1]. Fundus autofluorescence shows hyperautofluorescence of the vitelliform material and hypoautofluorescence in the atrophic stages [1]. In ARB, the marked autofluorescence changes, together with an absent EOG light rise disproportionate to the ERG, are the basis of the differential diagnosis [22]. Fluorescein angiography is difficult to interpret in BVMD, owing to dye accumulation within the lesion, staining linked to RPE pathology and the sub-RPE location of the neovascularisation; OCTA detects clinically unsuspected CNV and Han and colleagues recommend considering it in routine monitoring [56]. Ultra-widefield imaging reveals peripheral changes in BVMD and ARB [75]. In the IMPG1 and IMPG2 forms, the vitelliform material lies above the RPE, between the ellipsoid and interdigitation lines, with preserved RPE reflectivity, whereas in the BEST1 and PRPH2 forms the RPE complex participates in the lesion [6, 24]. Meunier and colleagues propose requesting IMPG1 and IMPG2 in the presence of moderate visual impairment, drusen-like lesions, an RPE line of normal reflectivity and deposits between the ellipsoid and interdigitation lines [24].

The French PNDS gives autofluorescence a central role: it considers it indispensable for diagnosis and holds that it has replaced fluorescein angiography, which is no longer indicated in the investigation of hereditary dystrophies [21]. This position differs from the practice described in GeneReviews and by Han and colleagues, who use leakage on angiography to characterise exudative CNV (section 7.3) [1, 56]. The PNDS proposes a decision tree based on the autofluorescence pattern: unilateral or asymmetrical abnormalities point to the vitelliform dystrophies, including Best disease; in bilateral and symmetrical abnormalities, round deposits point to the bestrophinopathies and pattern dystrophies, flavimaculate flecks to Stargardt disease, pseudo-Stargardt forms, pattern dystrophies and mitochondrial disorders, a perifoveal hyperautofluorescent ring to retinitis pigmentosa, cone or cone–rod dystrophies and maculopathies such as Stargardt disease, a fovea that is «too wide and too visible» to cone or cone–rod dystrophies and Stargardt disease, a granular appearance to cone dystrophies, achromatopsia and Stargardt disease, and cystic elements to the macular oedema of retinitis pigmentosa, retinoschisis, Goldmann–Favre disease and CRB1 maculopathy [21]. The initial examination is adapted to age: in young children, rapid-acquisition OCT scans, macular and if possible temporal, are preferred, so as not to miss an incipient mixed dystrophy, together with the full-field ERG; in older children and adults, visual acuity, visual field, colour fundus photographs, SD-OCT with macular and temporal scans and study of the ganglion cell fibre layer, macular and widefield autofluorescence, full-field ERG and, if the ERG is normal in a deposit disease or a toxic cause is suspected, EOG are recommended [21].

5.5. Genetic testing

GeneReviews proposes two approaches [1]. When the phenotype is suggestive, BEST1 sequencing is performed and, if negative, deletion and duplication analysis; in patients of Swedish ancestry the founder variant p.(Trp93Cys) can be tested first [1]. Alternatively, a multigene retinal dystrophy panel is used that includes BEST1 and the genes of the differential diagnosis (PRPH2, IMPG1, IMPG2, ABCA4 and others), bearing in mind that the genes included and the sensitivity vary between laboratories [1]. When the diagnosis has not been considered, the exome (more widely used) or the genome allows the broad approach [1]. The AAO recommends genetic testing for most patients with presumed inherited retinal degeneration, at the initial visit, with repetition during follow-up when the result is inconclusive, and considers it necessary for eligibility for gene therapy trials [85].

In France, the PNDS provides for next-generation sequencing of a retinal dystrophy gene panel (208 genes at the time of publication), with exon-level copy-number analysis, interpreted according to the ACMG/AMP criteria, and study of the parents' DNA in recessive forms to confirm by segregation that the two variants come from different parents [21]. In the France Genomic Medicine Plan 2025 (PFMG2025), maculopathies became a pre-indication for genome sequencing, which can be offered first line, preferably as a trio with the parents; the PNDS stresses that the pre-indications and trio criteria may change [21]. The same document counts 21 hereditary maculopathy genes, excluding the cone dystrophies, and describes BEST1 variants as mainly missense in the first half of the gene, some associated with the dominant forms and others with the recessive forms, the latter requiring a second variant in trans [21].

The proportion in the table refers to the variants detected by each method among probands in whom a variant was found and not to the clinical sensitivity of the test in the presence of a vitelliform phenotype [1]. In practice, the yield depends on the starting phenotype. In the Israeli consortium, only 76 of 134 patients (57%) with a clinical or electrophysiological diagnosis of BVMD or related conditions had BEST1 variants; in 9 families no variant was found and, in 4 families with a BVMD-like phenotype, the exome identified variants in PRPH2 and IMPG2 [20]. In adult-onset vitelliform macular dystrophy, Felbor and colleagues found PRPH2 variants in 18% of patients, Krämer and colleagues BEST1 variants in 4 of 32 patients, and Meunier and colleagues IMPG1 or IMPG2 variants in 8% of families already negative for BEST1 and PRPH2 [8, 24, 43]. The overall yield of genetic testing in inherited retinal dystrophies lies between 56% and 76% [85]. The figures of the French PNDS, without indication of the source series, are of the same order: a positive result in more than half of patients with maculopathy, in about 70% of pseudovitelliform dystrophies, in more than 62% of cases of Stargardt disease and in about 30% of recessive maculopathies other than Stargardt disease [21]. A negative result may be due to variants in unexplored regions of the panel genes, in genes outside the panel or in genes not yet described, which justifies the exome or the genome; a variant in an unexpected gene should prompt review of the clinical diagnosis, and novel variants may require functional studies to be classified [21].

The limitations of the test should be stated in the interpretation. The exome, which detected most variants in the Israeli cohort, does not exclude large deletions or insertions, and 9 of 56 index families remained without an identified variant [20]. Analysis of leucocyte DNA may not detect somatic mosaicism [1]. A negative result does not exclude the disease, and negative or inconclusive tests should be reviewed as new genes and reclassifications emerge [85]. In ARB, detection of a single variant (as in one of the 10 patients of the Boon series) requires a search for the second allele by CNV analysis or analysis of non-coding regions and confirmation of phase by parental testing [1, 22]. Apparent homozygosity with only one carrier parent suggests a large deletion on the other allele or uniparental isodisomy of chromosome 11 [1]. In IMPG2, the number of alleles (one or two) defines the disease, the prognosis and the mode of inheritance, and the report should state it explicitly [11]. The report should give the gene, the transcript with version (NM_004183.4 for BEST1), the HGVS c. and p. nomenclature, the zygosity, the class according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) and, for recessive variants, the phase when known [1]. Published notations that do not correspond to the current transcript, such as those listed in section 2.4, should be normalised with a validation tool before being used in reports or in cascade testing [7].

5.6. Practical algorithm

In the presence of a vitelliform lesion, the first step is multimodal imaging (fundus photography, SD-OCT, autofluorescence and, if CNV is suspected, OCTA), which defines the topography (macular, multifocal, peripheral) and the position of the material relative to the RPE [1, 24, 56]. The second step is electrophysiology: an EOG with a reduced LP:DT ratio and a normal full-field ERG point to BVMD; a disproportionately reduced EOG with an abnormal ERG points to ARB or ADVIRC; a normal or borderline EOG in an adult vitelliform lesion points to non-BEST1 AFVD, particularly IMPG1/IMPG2 or PRPH2 [24, 25]. In patients with ARB, BVMD or ADVIRC, the anterior chamber angle should be assessed by gonioscopy and biometry [29, 60]. The third step is genetic testing with a retinal dystrophy panel, with CNV analysis, and parental testing for segregation and phase [1, 85]. A VUS should not guide predictive testing of relatives; family segregation and periodic reassessment may reclassify it [1, 85].

5.7. Prenatal and preimplantation diagnosis

Once the familial variants have been identified, prenatal diagnosis and preimplantation genetic testing are technically possible; intrafamilial variability prevents prediction of expression and age at onset from the result [1]. In France, the PNDS places the request for prenatal or preimplantation diagnosis within a consultation with a geneticist or genetic counsellor, accepts it only very restrictively and case by case, and stresses the limitation imposed by the difficulty of distinguishing rare non-pathogenic variants from pathogenic variants [21]. Counselling aspects are covered in section 8.

6. Differential Diagnosis

6.1. Comparative table

AD, autosomal dominant; AR, autosomal recessive.

6.2. Main differentials

Sorsby fundus dystrophy is also among the hereditary macular dystrophies to be considered [62]. The most frequent distinction in children is from Stargardt disease, which GeneReviews gives as the most frequent juvenile hereditary macular dystrophy, with BVMD in second place; linear yellow flecks, macular atrophy and an EOG proportional to the ERG point to ABCA4 [1]. In adults, a small, bilateral vitelliform lesion raises the differential between BEST1-associated AVMD, AFVD due to PRPH2, IMPG1 or IMPG2, pattern dystrophies and AMD; most cases of AFVD remain without an identified cause and the EOG, the position of the vitelliform material relative to the RPE and the genetic panel are the discriminators [1, 24, 25]. Krämer and colleagues found no BEST1 variants in 200 patients with AMD, nor any association of the locus with AMD [8]. Cataract surgery in eyes with AFVD (18 eyes) produced an improvement in visual acuity from 0.60 ± 0.34 to 0.23 ± 0.11 logMAR at one month, without new CNV or atrophy, similar to that observed in eyes with AMD [31].

ARB and ADVIRC are confused with retinitis pigmentosa, cone–rod dystrophies, retinoschisis and primary angle-closure glaucoma; the combination of a short eye, a closed angle and unexplained retinal changes in a young adult should prompt study of BEST1 [29, 59]. Recently described differentials include THRB-associated macular dystrophy, in which a heterozygous splicing variant c.283+1G>A produced, in one family, a picture of occult macular dystrophy in the mother and vitelliform dystrophy in the son [91], retinal vitelliform lesions as an adult-onset presentation of the mitochondrial variant m.3243A>G [92], and the coexistence of BEST1 and EFEMP1 variants in the same patient [89]. Occult macular dystrophy, whose picture the mother in the THRB family presented, has as its classical cause autosomal dominant missense variants in RP1L1, the most common of which is NM_178857.6:c.133C>T p.(Arg45Trp); in a French multicentre cohort of 20 patients from 19 families with RP1L1-associated retinopathy, 12 had occult macular dystrophy, 9 of them with p.(Arg45Trp), the mean age at onset was 31.5 years (6 to 66), the median BCVA was 0.6 logMAR and declined by about one ETDRS line every two years, and the penetrance of the dominant forms was estimated at 85% [90]. It is distinguished from BVMD and AVMD by the absence of vitelliform material and by normal or near-normal autofluorescence, the multifocal ERG being the test that demonstrates macular dysfunction; the ffERG was normal in most patients, but one patient homozygous for p.(Arg45Trp) had generalised cone–rod dysfunction with onset at 9 years of age, so the ffERG should be performed even when the picture appears purely macular [90]. Acute exudative polymorphous vitelliform maculopathy, in its idiopathic and paraneoplastic forms, produces multiple vitelliform lesions without a genetic basis [93].

The French PNDS also lists, among the rare hereditary maculopathies, the dominant form of Stargardt disease due to ELOVL4, the PROM1, CRX, CRB1 and RP1 maculopathies, MIDD, Sorsby dystrophy, Malattia leventinese and North Carolina dystrophy, and stresses that the list is not exhaustive and that it is genetic testing that separates these entities [21]. The distinction from the cone dystrophies rests on early photophobia, colour vision changes that are more severe for the same level of acuity and the full-field ERG, with reduced or extinguished cone responses and preserved rod responses [21]. After 50 years of age, AMD is distinguished by more asymmetrical involvement, one eye after the other, and extensive macular atrophy with pseudodrusen by its rapid progression to legal blindness; dome-shaped macula and occult maculopathies keep a normal full-field ERG and require macular OCT and multifocal ERG [21].

7. Treatments and Risk Management Strategies

7.1. General principles and framework of the recommendations

There is no treatment approved by the US Food and Drug Administration (FDA) or by the European Medicines Agency (EMA) for any bestrophinopathy [23, 94]. Nor is there a specific clinical practice guideline: GeneReviews records the absence of published guidelines in 2020, the European network ERN-EYE has no consensus document on Best disease (its consensus section contains only the one on Bardet–Biedl, from 2024) and the guidelines applicable to assessment and follow-up are the 2022 AAO clinical statement on inherited retinal degenerations, in which the macular dystrophies constitute class 4, and the French National Diagnosis and Care Protocol (PNDS) «Maculopathies génétiques», published by the Haute Autorité de Santé (HAS) in July 2021, prepared by the MAOLYA reference centre (Montpellier) within the SENSGENE network and reviewed by the CARGO and REFERET centres, which covers Stargardt disease, Best disease and the PRPH2 maculopathies together [1, 21, 32, 85]. The PNDS is based on a review of the literature published between 1970 and 2020 and on the experience of the reference centres, without grades of recommendation, and does not give drug doses [21]. The 2017 ISCEV standard regulates the EOG [25]. Current treatment is directed at complications (CNV, angle-closure glaucoma, amblyopia, macular oedema), visual rehabilitation and genetic counselling, and gene therapy is in an early-phase trial [1, 28].

After diagnosis, GeneReviews recommends an ophthalmological examination with best-corrected visual acuity (BCVA), fundoscopy, fundus photographs and SD-OCT for staging, a low-vision consultation as needed and a genetics clinic consultation [1]. The AAO adds, at the initial visit, a history that includes retinotoxic drugs, a pedigree, BCVA on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart, intraocular pressure (IOP), slit-lamp biomicroscopy, autofluorescence and genetic testing, and a low-vision assessment at baseline [85]. The French PNDS sets as treatment objectives the management of low vision, treatment of complications, general lifestyle measures and overall support for the patient and family, and assigns four functions to the family doctor and the ophthalmologist: to suspect the diagnosis and refer to a reference or competence centre, to ensure regular follow-up by teams trained in visual impairment and rehabilitation, to coordinate the medico-social aspects (100% coverage and an application to the Maison départementale des personnes handicapées, MDPH) and to direct the patient towards appropriate educational and rehabilitation services [21].

7.2. Visual rehabilitation, education and amblyopia

Low-vision aids are indicated when visual acuity is significantly reduced [1]. In the United States, GeneReviews describes the individualised education programme (IEP, reviewed annually, with the least restrictive environment possible and a vision consultant on the team) and the 504 plan (front-row seating in class, assistive technology, a scribe, extra time, adapted assignments, enlarged text), as well as occupational counselling [1]. The sources consulted do not describe the equivalent mechanisms in Portugal or Brazil. In children, GeneReviews recommends annual examinations to prevent amblyopia, particularly when there is a significant difference in BCVA between the eyes, and a conventional occlusion trial of the better eye can clarify whether amblyopia is present [1]. In the family described by Iannaccone and colleagues, treatment of amblyopia in the eye with a subfoveal vitelliform scar improved visual acuity from 20/200 to 20/50 [2].

The French PNDS adds concrete measures. In Best disease, the frequent hyperopia should be corrected as early as possible to reduce the risk of amblyopia, and all patients should have the best possible optical correction [21]. Early low-vision rehabilitation is led by a specialised orthoptist, in liaison with the ophthalmologist and the low-vision optician, close to home or in a multidisciplinary centre; therapeutic education focuses on training eccentric fixation, which teaches use of the unaffected retina, as a rule above the fovea, and on the use of optical aids (magnifying spectacles, magnifiers, telescopes), non-optical aids (large-print books, talking clocks, enlarged telephone keypads) and electronic aids (reading systems); the team may intervene at home to apply safety rules and participate in adapting the workstation [21]. For photophobia, contrast-enhancing filter lenses are indicated, with a wraparound frame with wide arms; there are mini-cameras attached to spectacles that read text pointed at with a finger, and voice or character recognition software; Braille and institutional care are rarely needed in these maculopathies, and a mobility instructor and a guide dog can help in severe impairment [21]. The occupational therapist is useful mainly in the syndromic forms, physiotherapy treats the musculoskeletal pain caused by the postures adopted in order to see, and psychological support should be offered to the patient and family, because the phase of falling acuity is often accompanied by anxiety and depression, which tend to ease when acuity stabilises and the patient learns to use the preserved retina [21].

At school, a child with a non-syndromic maculopathy follows mainstream education, with a seat facing the board, adequate lighting, a computer with magnification and audio software, a tablet to photograph the board and documents, materials in digital format, a school support assistant when necessary and extra time in tests and examinations, which in France may be formalised in an individualised reception plan (PAI) [21]. In adults, the PNDS states that, in France, driving is no longer permitted with a visual acuity below 5/10 and/or a central scotoma, and provides for vocational reorientation with the occupational physician and the social worker and, when necessary, comprehensive rehabilitation in specialised institutes [21]. The legal criteria for fitness to drive in Portugal and Brazil are not given in the sources consulted.

7.3. Choroidal neovascularisation: when to treat

The decision to treat CNV in BVMD is not based on trials. The two OCTA series converge on the distinction between exudative and non-exudative neovascularisation, but diverge on the treatment threshold. Parodi and colleagues treated only the exudative neovascularisation of stages 2–3 (4 of 30 eyes), with ranibizumab and clinical stabilisation, and observed without treatment the 46 non-exudative neovascularisations of stages 4–5, which remained stable [76]. Han and colleagues observe non-exudative CNV in stable patients, but offer a therapeutic trial of anti-VEGF not only in the presence of obvious new exudation (intraretinal or subretinal haemorrhage, clear leakage on fluorescein angiography), but also when exudation is suspected, defined as subjective visual loss with a reduction of the subretinal fluid within the vitelliform lesion (vitelliform → vitelliruptive transition), even without definite CNV on imaging; they stop when function returns to baseline, when the CNV is no longer visible on OCTA or when the OCT reaches a plateau, extending the intervals as in inactive AMD [56]. In the longitudinal subgroup of this series, two patients with lesion collapse and CNV were treated and the others with CNV without visual change were only observed [56]. Both positions are author opinion based on single-centre series, and an evidence-based review of anti-VEGF agents in paediatric retinal diseases classified the literature on Best disease as consisting of case reports and retrospective series, of low level of evidence [95]. The French PNDS merely states that intravitreal anti-VEGF injections are to be discussed in the active phase of CNV and that they are neither indicated nor effective in the fibrotic phase, without defining activity criteria or a dosing schedule [21].

Before the anti-VEGF era, direct laser photocoagulation was used in stage 4c or in CNV with haemorrhage, while Marano and colleagues suggested a conservative approach on the basis of two patients with BVMD whose visual acuity improved spontaneously; there are no trials comparing laser and observation [1, 96]. GeneReviews considers anti-VEGF agents the standard treatment of subfoveal CNV in general and cites intravitreal bevacizumab in a 13-year-old boy with BVMD, with faster visual recovery and regression [97], successful intravitreal ranibizumab [98] and the absence, at the time, of reports with aflibercept; long-term follow-up is unknown [1]. Photodynamic therapy with verteporfin produced regression of CNV and resolution of subretinal haemorrhage in one adult patient [99] and, in an Italian series of 30 patients with BEST1 variants, 5 had CNV before 15 years of age and 3 were treated with photodynamic therapy, with stable visual acuity in the first year and slow improvement thereafter, associated with macular fibrosis, over a mean follow-up of 77 months (62–99) and without adverse effects [100]. The verteporfin dosage used in these reports is not given in the sources consulted.

7.4. Anti-VEGF drugs: dosage and safety

The intravitreal anti-VEGF agents used in CNV of the bestrophinopathies are ranibizumab, aflibercept and bevacizumab. No label mentions Best disease; the doses below are those of the labels in the closest indications, or those described in case reports, and the regulatory framework differs between jurisdictions (section 7.5).

The sources consulted do not include the lactation section of the labels or lactation data in the bestrophinopathies.

7.5. Comparative regulatory status

Sources: official FDA and EMA labels and lists [26, 27, 101–103]; OPGx-BEST1 according to the sponsor's quarterly report and press release [94, 105]. The sources consulted include no data on reimbursement or national access in Portugal or Brazil for the use of these drugs in Best disease.

7.6. Trials and series supporting treatment

Interpretation of this evidence has limitations that constrain practice. No randomised trial included patients with bestrophinopathy: the efficacy of ranibizumab in MINERVA was demonstrated in other CNV aetiologies and its extrapolation to Best disease is indirect, although aetiology did not appear to modify the direction of the effect [26]. The Han and Parodi series do not compare treatment with observation, and indirect comparison between them is limited by differences in design (retrospective versus prospective), age and definition of activity [56, 76]. The high frequency of stable non-exudative neovascularisation indicates that detection of CNV on OCTA, in isolation, is not an indication for treatment [56, 76]. The gene therapy data come from a press release and a report to the Securities and Exchange Commission (SEC), without peer review, without a control group, with five patients and three to six months of follow-up; «clinically meaningful improvement» is defined by the sponsor and no magnitudes or confidence intervals have been disclosed [94, 105]. A search of ClinicalTrials.gov in October 2026 found no suspended, terminated or negative BEST1-specific interventional trials; the absence of records does not prove that none exist [28].

7.7. Angle-closure glaucoma

Angle-closure glaucoma is the complication that most determines the visual prognosis in ARB: in the French cohort, patients with angle closure reached severe visual impairment at a median age of 47 years, against 68 years in the remainder [60]. Crowley and colleagues, reporting a young man with homozygous c.636+1G>A ARB treated with topical therapy, peripheral iridotomies and phacoemulsification, recommend that all patients with ARB or BVMD be examined for narrow angles and glaucoma [29]. Boon and colleagues, in the Dutch series, state that several currently available treatment options should be considered; the detail of those options appears in the full text, which was not accessible [22].

Filtering surgery in these short eyes carries its own risks. In a series of three patients with ARB and angle-closure glaucoma, aged 35 to 48 years, with preoperative IOP of 27, 28 and 47 mmHg, axial length of 22.73, 21.52 and 20.96 mm and anterior chamber depth of 2.51, 1.97 and 2.19 mm, all three developed malignant glaucoma immediately after trabeculectomy, resolved by a second operation that included anterior vitrectomy and posterior capsulotomy [30]. Alternatives are described in isolated cases: in a 29-year-old woman with glaucoma refractory to repeated trabeculectomies, ultrasound cyclodestruction in the left eye stabilised the IOP and reduced the schisis-like changes and the subretinal fluid at 7 months, while the fellow eye on medical treatment progressed [59]; in a 12-year-old girl, micropulse transscleral cyclophotocoagulation deepened the anterior chamber, lowered the IOP and resolved the retinoschisis, with partial regression of the deepening from day 12 and recurrence of the schisis at day 67 [81]; in a 28-year-old man with homozygous ARB and bilateral angle closure, a central retinal vein occlusion in the left eye was treated with intravitreal anti-VEGF, panretinal photocoagulation and trabeculectomy, followed by trabeculectomy in the right eye, with IOP of 12 mmHg or less without medication at one year [80]. These are case reports, without comparison between techniques. The sources consulted include no specific recommendations on pharmacological mydriasis in patients with ARB and a narrow angle; assessment of the angle before dilation follows from the general recommendation of Crowley and colleagues [29].

7.8. Macular oedema, cataract and other interventions

Cystoid macular oedema occurs in ARB and in ADVIRC [1]. The sources consulted do not describe treatment of macular oedema specific to the bestrophinopathies; in the retinitis pigmentosa phenotypes associated with PRPH2 and IMPG2, the GeneGuidance curation indicates carbonic anhydrase inhibitors (oral acetazolamide, topical dorzolamide), with a possible rebound effect [10, 11]. Cataract surgery in eyes with AFVD produced improvement in visual acuity without new CNV or atrophy at one month, in a series of 18 eyes [31]. For the retinitis pigmentosa phenotypes, the same curation recommends avoiding high-dose vitamin E (for example, 400 IU/day), bright light and smoking [10]. There is no gene therapy for PRPH2 [10].

7.9. Gene therapy and gene editing: developments from October 2025 to October 2026

The biological rationale for gene therapy in the bestrophinopathies rests on three arguments: slow progression with prolonged preservation of the central photoreceptors, which offers a therapeutic window; the existence of structural outcomes quantifiable by OCT and autofluorescence (subretinal fluid, vitelliform lesions); and the possibility of choosing the strategy according to the mechanism of the variant [23]. Loss-of-function and dominant negative variants are candidates for gene augmentation; gain-of-function variants require silencing of the endogenous allele combined with replacement [23, 46].

Preclinical evidence. In the canine model, subretinal gene augmentation with an adeno-associated virus serotype 2 (AAV2) vector reversed the subretinal lesions and microdetachments in three canine genotypes, with titres of 0.1 to 5×10¹¹ vg/mL [45]. The good laboratory practice toxicity and efficacy study of OPGx-BEST1 (rAAV2/2-VMD2-BEST1: human BEST1 cDNA under the VMD2/BEST1 promoter, AAV2 capsid) treated 12 mutant dogs, in four groups of three, with a subretinal injection of 0.15 mL in the left eye at 9.5×10⁹, 3.0×10¹⁰ or 3.0×10¹¹ vg/mL (doses of 1.4×10⁹, 4.5×10⁹ and 4.5×10¹⁰ vg) or vehicle; there was no systemic or ocular toxicity, rod and cone responses on ERG improved and OCT showed reduction or prevention of the vitelliform lesions and reversal of the microdetachments in the treated areas, with a no observed adverse effect level of 4.5×10¹⁰ vg/eye [106]. In RPE derived from human iPSC, gene augmentation restored the calcium-activated chloride current and rhodopsin degradation in a model of ARB and in two of three models of dominant BVMD with variants in ion-binding domains; the third did not respond, but was normalised with CRISPR-Cas9 editing specific to the mutant allele, which led the authors to propose gene augmentation as first line for some patients and editing for non-responders [107]. In patient-derived RPE, supplementation of wild-type BEST1 by AAV restored the calcium-dependent chloride currents with similar efficacy for dominant and recessive variants [48]. Gain-of-function variants were not rescued by gene augmentation alone, but were rescued by its combination with silencing of endogenous BEST1 by dCas9-KRAB-MeCP2 [46].

Gene editing. In 2026, Milenkovic and Weber tested allele-specific guide RNAs for p.(Arg218Cys), p.(Ala243Val) and p.(Ile295del) in iPSC-derived RPE; only one of the three variants obtained a guide with high cleavage efficiency and no detectable off-target activity, and elimination of the mutant transcript improved the localisation and stability of bestrophin-1 and restored anion transport, which shows that allele-specific editing is feasible for selected variants but requires each guide to be tested [54]. A July 2026 preprint, without peer review, describes correction of the variant c.851A>G p.(Tyr284Cys) by base editing (cytidine) in iPSC, with normalisation of the monolayer, the barrier and the RPE transcriptome, and delivery by dual AAV with intein splitting to mutant monolayers, with a maximum editing efficiency of 13.42 ± 3.64% [57].

BIRD-1 clinical trial (OPGx-BEST1). Trial NCT07185256 (DUO-1001), from Opus Genetics, is a phase 1b/2a, open-label, dose-escalation, basket-type study, with a single subretinal injection in one eye of adults (18 years or older) with BVMD or ARB; the vector contains the full codon-optimised human BEST1 under the RPE-specific VMD2 promoter [28]. It began on 25 September 2025, plans 10 participants and completion in August 2030, and was recruiting at the last update of 24 September 2026, in five centres in the USA [28]. The primary outcomes are dose-limiting toxicities (grade 3 or higher within 90 days of administration), adverse events related to the procedure (pars plana vitrectomy and subretinal injection under general anaesthesia) and those related to the drug, up to 5 years; the secondary outcomes include SD-OCT, autofluorescence, OCTA, microperimetry, ETDRS BCVA, low-luminance visual acuity and the Michigan Retinal Degeneration questionnaire [28]. The sentinel participant in each cohort has a BCVA of 20 letters or fewer (20/200 or worse) and those following between 65 and 20 letters (20/50 to 20/400) [28]. The FDA accepted the investigational new drug (IND) application in August 2025 and the first participant was treated in November 2025; the doses are 1.5×10⁹ vg/eye in cohort 1 and 4.5×10⁹ vg/eye in cohort 2 [105, 108]. In February 2026, the sentinel data presented at the Macula Society indicated absence of ocular inflammation, of treatment-related adverse events and of dose-limiting toxicity, a gain equivalent to 12 BCVA letters and a 23% reduction in central subfield thickness in the treated eye [105]. On 9 September 2026, the sponsor reported the results of cohort 1 (five participants: three with BVMD followed for 3 months and two with ARB followed for 6 months): no serious adverse events, dose-limiting toxicity or intraocular inflammation, with all treatment-related events mild or moderate; improvement in at least one functional measure in all five, concentrated in the «transition zone» of the treated RPE and greater in less advanced disease; and structural improvement in four of five (reduction of vitelliform material in two of the three patients with BVMD and of intraretinal fluid in the two patients with ARB) [94]. On the same date, the sponsor reported a type C meeting with the FDA, in August 2026, at which an improvement of at least 3 dB on microperimetry in at least 5 prespecified loci, combined with a patient-reported outcome, was discussed as a potential primary endpoint for phase 3, in a randomised controlled trial, with dosing in phase 3 expected to begin in 2027; cohort 2 was expanded from 5 to 8 participants [94]. These data have the evidence level of a company announcement (8-K and 10-Q filings with the SEC), reproduced by specialist press that does not constitute independent confirmation [94, 109, 110]. The quarterly report does not mention orphan drug designation or other designations for OPGx-BEST1, which does not demonstrate that none exist [105]. The Foundation Fighting Blindness funded the proof-of-concept studies at the University of Pennsylvania with almost 2 million dollars, and its venture philanthropy arm founded Opus Genetics in 2021 [108].

Natural history studies. Trial NCT05809635, observational and prospective, coordinated by Columbia University with centres in Tübingen, Paris (Quinze-Vingts) and at the National Eye Institute, plans 52 patients with BVMD followed for up to 3 years, with dark-adapted chromatic perimetry, ERG and EOG according to ISCEV, OCT and quantitative autofluorescence [111]. The proposal to use the loss of sensitivity at initially normal microperimetry points as an outcome for gene therapy trials follows from the Italian natural history cohort [73].

7.10. Measures to avoid, special populations and transition

GeneReviews recommends smoking cessation to help prevent neovascularisation, on the basis of evidence derived from AMD (AREDS study, report 19) [1, 112]. The sources consulted do not identify specific drugs to be avoided in the bestrophinopathies; the AAO recommends recording exposure to retinotoxic drugs in the history [85]. The French PNDS recommends, for all genetic maculopathies, abstinence from smoking, a varied and balanced diet rich in fruit, vegetables and antioxidants, with fish three times a week, regular physical activity, even if moderate (about one hour a day), and protection against light (filtering lenses, a peaked hat, blue light filters or adjustment of screens) [21]. These are general preventive measures, without trials in the bestrophinopathies. The contraindication of vitamin A supplementation, the avoidance of quinine derivatives, the mandatory nature of light protection and the avoidance of combat sports refer in the PNDS to Stargardt disease due to ABCA4, because of the mechanism specific to the visual cycle and the risk of retinal haemorrhage, and are not extrapolated to Best disease [21].

In pregnancy there are no data specific to the bestrophinopathies. The North American label for ranibizumab states that it should be used in a pregnant woman only if clearly needed, following skeletal abnormalities in monkeys at exposures 13 times the human level, and that for aflibercept that it should be used only if the potential benefit justifies the risk to the fetus, following malformations in rabbits with no fetal no observed adverse effect level identified [27, 101]. In children, the safety and efficacy of ranibizumab are not established outside retinopathy of prematurity, and aflibercept has a paediatric indication only in that disease; the paediatric cases of BVMD and ARB treated with bevacizumab or photodynamic therapy are isolated reports [2, 26, 100, 101, 104]. Paediatric examination may require sedation for reliable results before the age of 2 years, and visual fields are unreliable before the age of 7 years [85]. BIRD-1 includes adults only [28].

In the perioperative context, eyes with ARB and ADVIRC, short and with a shallow anterior chamber, carry a high risk of malignant glaucoma after filtering surgery [30]. The subretinal injection in the BIRD-1 trial requires vitrectomy under general anaesthesia [28]. The French PNDS provides, at the end of adolescence, for an organised transfer to adult services, to avoid interruption of follow-up, and includes among the aims of follow-up the assessment of fitness to drive and for occupational activity [21]; the AAO recommends repeating genetic counselling later in life in early-onset diseases [85].

7.11. Recommendations of the reference sources and divergences

Sources: [1, 21, 25, 29, 56, 76, 85].

8. Reproductive Options and Family Counselling

8.1. Dominant forms: BVMD, AVMD and ADVIRC

Most patients with BVMD, AVMD or ADVIRC have an affected parent, but de novo cases have been described, in an unknown proportion [1]. GeneReviews recommends molecular testing of the parents of an apparently isolated proband, because fundoscopy may be unreliable: a heterozygote may have a normal fundus [1]. An apparently negative family history may result from failure to recognise the disease in relatives, from incomplete penetrance or from mosaicism; parental germline mosaicism is theoretically possible but has not been described, and leucocyte DNA may fail to detect somatic mosaicism [1]. The father of a patient with Best disease who had a circinate vitelliform dystrophy of the mid-periphery with a spared macula illustrates the variable expressivity that can mask transmission [61]. The French PNDS assigns to the EOG, together with testing for the familial variant, the role of recognising asymptomatic carriers, and points out that apparently isolated cases due to variants in dominant genes are frequent and carry a risk of transmission to the next generation, even when expression in the parent is mild or absent [21].

If a parent is affected or a carrier, the risk to each sib of the proband is 50%, with age at onset, manifestations and degree of disability varying widely among heterozygotes [1]. If the variant has not been detected in the parents, the risk to sibs is about 1%, owing to the possibility of germline mosaicism [1]. If the parents have not been tested but are clinically unaffected, with a normal EOG, the risk is low but higher than that of the general population because of incomplete penetrance and mosaicism [1]. Each child of a heterozygote has a 50% risk of inheriting the variant [1]. Penetrance above 70% means that not all heterozygotes will develop clinically evident disease [1].

8.2. Autosomal recessive bestrophinopathy

The parents of a patient with ARB are usually heterozygotes [1]. When only one parent is a carrier, apparent homozygosity suggests a large deletion on the other allele or uniparental isodisomy of chromosome 11, and compound heterozygosity suggests a de novo variant [1]. Sibs have a 25% risk of being affected, a 50% risk of being asymptomatic carriers and a 25% chance of being neither affected nor carriers; the children of a patient with ARB are obligate heterozygotes, and the sibs of the parents have a 50% risk of being carriers [1]. Carriers generally have no clinical disease and no electrophysiological abnormalities, with a normal EOG, which distinguishes ARB from BVMD, in which heterozygotes with a normal fundus may have an abnormal EOG [2, 3, 29]. Some carriers have, however, had a mildly reduced mfERG or mild symptoms of maculopathy [1, 63, 64]. In the Israeli cohort, consanguinity was 42% among Muslim Arabs and 13.6% or less among Jews, and 14 of the 43 families with variants had recessive disease [20]. The distinction between BVMD with a dominant variant and ARB has direct consequences: the same phenotype of multiple vitelliform lesions may correspond to a 50% risk to children (dominant) or to a low risk to children and a 25% risk to sibs (recessive) [1, 2].

8.3. Non-BEST1 vitelliform dystrophies

IMPG2 heterozygotes may be asymptomatic with subtle foveal irregularities, so an apparently unaffected parent does not exclude dominant transmission; in IMPG2 and IMPG1, the allele count defines the mode of inheritance and the recurrence risk [11–13]. In sporadic cases of undiagnosed macular dystrophy, the Oxford Desk Reference considers a genetic aetiology likely and the recurrence risk high, with recessive inheritance the most likely, but notes that dominant vitelliform dystrophies have incomplete penetrance and variable expressivity, and recommends storing DNA [62].

8.4. Cascade testing, minors and reproductive planning

Carrier detection and predictive testing require prior knowledge of the familial variants [1]. GeneReviews recommends clarifying the genetic status of asymptomatic relatives, both older and younger, to allow early ophthalmological assessment and follow-up, including prevention of amblyopia in children [1]. The AAO stresses that testing asymptomatic relatives for diseases without established treatment requires genetic counselling, that results should be interpreted and communicated by an experienced physician or a genetic counsellor, and that counselling should be repeated later in life in early-onset diseases [85]. The optimal time for reproductive counselling is before pregnancy, and it should be offered to young adults who are affected, carriers or at risk [1]. Prenatal diagnosis and preimplantation genetic testing are possible once the familial variants have been identified, but intrafamilial variability does not allow the age at onset or severity to be predicted from the result [1]. Genetic testing is also required for eligibility for gene therapy trials [21, 85].

The French PNDS organises the disclosure of the diagnosis in a dedicated consultation, possibly with several members of the team (ophthalmologist, geneticist, paediatrician, psychologist, orthoptist, social worker), which explains the disease, its course and the existing treatments or their absence, introduces the follow-up team and patient associations and starts genetic counselling; the genetics clinic consultation should take place as early as possible, and support for the parents and for local professionals continues at each difficult stage [21]. Genetic testing requires written consent; in a minor, it is performed only with the agreement of the parents or guardian and when clinical signs are present, so that, in a dominant disease, an unaffected minor cannot be tested to learn his or her genetic status [21]. This rule differs from the GeneReviews recommendation to clarify the genetic status of younger relatives to guide surveillance (section 7.11); the PNDS further requires discussion with the parents of the psychological impact of systematic clinical examination of the sibs of young patients [1, 21].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

Prevalence estimates for the bestrophinopathies differ according to the numerator (clinical diagnosis, genetically confirmed diagnosis, all vitelliform dystrophies or all BEST1 retinopathies), the denominator and the method of ascertainment.

The figure of 1:5500, presented by GeneReviews and by the GeneGuidance curation as the prevalence of BVMD in a North American population, corresponds in the original article to the prevalence of all vitelliform dystrophies in Olmsted County; the prevalence of Best disease in the same study is 1:16,500 to 1:21,000 [1, 19, 44]. The review by Haldrup and colleagues cites a range of 1:5000 to 1:67,000 for BVMD and a prevalence of 1:1,000,000 for ADVIRC [23]. In ARB, Orphanet counts 20 cases or families, but recent series are larger (34 patients in Paris, 74 in China, 15 in Norway, 14 families in Israel) [15, 20, 55, 60, 72]. In ADVIRC, Orphanet counts 3 families and the Norwegian series included 4 patients [16, 55]. In Norway, BEST1 variants are the third most common cause of inherited retinal dystrophy [55].

The commercial estimates of the sponsor of OPGx-BEST1, obtained by surveying more than 150 eye care professionals, of about 23,600 symptomatic patients in the USA (13,000 diagnosed) and 45,400 worldwide, or the earlier estimate of 8400 and 21,800, are not population-based epidemiological studies [94, 105].

9.2. Ancestry, founder effect and proportion of recessive forms

The p.(Trp93Cys) variant is a founder variant among the descendants of the large Swedish kindred known as «pedigree S1» [1, 36]. In Israel, 6 of the 20 variants were recurrent in families of the same origin, suggesting a founder effect, and novel variants associated with ARB were described: c.967G>A p.(Asp323Asn), c.970G>T (nonsense), c.1740-1G>C and c.1622del [20]. The proportion of recessive alleles was 38.3% in Israel, compared with 32.3% in the published worldwide total [20]. In Denmark, a variant at codon 302 was recurrent in four unrelated families [18]. In Norway, c.89A>G and c.403G>A predominate [55]. According to the preprint meta-analysis, biallelic patients would account for about 20% of the BEST1 population in Western countries and about one third in East Asia, and ARB would be reported as such less often in the USA than in Europe [113]. The Chinese cohort, with 74 patients with ARB and 60 with BVMD, is consistent with a higher proportion of recessive forms in East Asia [72].

9.3. Portugal, Brazil and gaps

The sources consulted do not include studies of the prevalence or variant spectrum of BEST1 in Portugal or Brazil, which does not demonstrate that no published data exist. The Portuguese member of the ERN-EYE network is the Unidade Local de Saúde de Coimbra [32].

9.4. Special populations

Children are the group in which BVMD usually presents and in which treatment of CNV is most often off-label (section 7.10) [1, 26]. Patients from populations with high consanguinity have a higher proportion of ARB [20]. Patients with ARB and short eyes constitute a group at surgical risk [30].

9.5. European network and patient associations

ERN-EYE (European Reference Network for Rare Eye Diseases; https://www.ern-eye.eu) brings together 52 full members and 8 affiliated partners in 24 of the 27 European Union countries, covers more than 900 rare eye diseases and runs a distance-learning programme on inherited retinal dystrophies [32]. Patient associations and resources include Retina International (https://retina-international.org), the Foundation Fighting Blindness (https://www.fightingblindness.org), with the My Retina Tracker registry, the Macular Society in the United Kingdom (https://www.macularsociety.org), PRO RETINA Deutschland (https://www.pro-retina.de), Retina UK (https://retinauk.org.uk) and, in Brazil, Retina Brasil (https://retinabrasil.org.br), with addresses confirmed in October 2026. The eyeGENE repository of the National Eye Institute collects samples and data from patients with inherited eye diseases, and GeneReviews also lists the Macular Degeneration Foundation and the Association for Macular Diseases [1]. No official website of a Portuguese association of patients with retinal dystrophies was identified in the searches performed.

In France, the organisation of care rests on the SENSGENE rare sensory diseases health network (https://www.sensgene.com) and on the rare disease reference centres (CRMR) for genetic ophthalmology: MAOLYA (genetic sensory disorders, CHU de Montpellier, coordinator of the PNDS), REFERET (retinal and macular disorders of genetic origin, Centre hospitalier national d'ophtalmologie des Quinze-Vingts, Paris), CARGO (rare disorders in ophthalmic genetics, CHU de Strasbourg) and OPHTARA (rare ophthalmological diseases, Hôpital Necker-Enfants malades, Paris), together with the reference centre for genetic deafness (relevant to MIDD) and regional competence centres in Clermont-Ferrand, Grenoble, Brest, Lille, Nantes, Marseille, Nice, Bordeaux, Poitiers, Toulouse and Saint-Denis (Réunion) [21]. The PNDS recommends that follow-up be coordinated by a specialist from a CRMR or competence centre, who organises the care network close to home; the disease may qualify for 100% coverage as a condition outside the list of long-term conditions (exemption from the ticket modérateur), and the MDPH assesses disability benefits, including support for guide dog applications [21]. The French patient associations listed in the PNDS include Retina France (http://www.retina.fr), the Fédération des Aveugles de France, the Association Valentin Haüy, RETGEN, Information Recherche Rétinite Pigmentaire, SOS Rétinite, the Fondation Stargardt and Histoire de voir un peu; referral to an association should always be offered, with the choice left to the patient [21].

10. Illustrative Clinical Case

The following case was published by Han and colleagues, from the University of Iowa, as patient 13 of the series of 27 patients with BVMD studied by swept-source OCTA [56]. A 27-year-old man, with BVMD molecularly confirmed by a heterozygous BEST1 variant p.(Lys30Arg), corresponding to NM_004183.4:c.89A>G, presented with one month of reduced visual acuity and progressive subjective distortion in the right eye [7, 56]. BCVA was 20/25 in both eyes. The symptomatic right eye had a vitelliruptive lesion (Gass stage 4) with minimal subretinal fluid on OCT, and OCTA showed CNV associated with a small pigment epithelial detachment [56].

The authors administered intravitreal bevacizumab 1.25 mg/0.05 mL at the initial visit and repeated the injection after one month because of persistent CNV [56]. At two months, BCVA had returned to 20/20, with resolution of symptoms; OCTA showed regression of the CNV and there was reaccumulation of subretinal fluid within the lesion, so further treatment was deferred [56]. At six months, BCVA remained 20/20 and the vitelliform lesion had reconstituted, with reaccumulation of subretinal fluid and no CNV visible on OCTA, that is, a regression from Gass stage 4 to stage 2 [56]. Figure 4 of the article, consulted on the PubMed Central figure page (https://pmc.ncbi.nlm.nih.gov/articles/PMC10164042/figure/F4/), documents this sequence, with the legend «Reversion of a vitelleruptive lesion (Gass stage 4) to a vitelliform lesion (Gass stage 2) after the treatment of choroidal neovascular membrane (CNV)» [56]. The authors describe a second similar case in the supplementary material [56].

The case illustrates three points with practical consequences. First, the transition from a vitelliform to a vitelliruptive lesion with recent visual loss may reflect underlying CNV, sometimes inconspicuous on fluorescein angiography and identifiable on OCTA [56]. Second, the reaccumulation of subretinal fluid after anti-VEGF did not indicate neovascular activity but the reconstitution of the vitelliform lesion, in keeping with the pathophysiological model proposed by the authors, in which CNV breaches the RPE and drains the fluid into the choroid [56]. Third, the management described is that of one centre and involves an off-label drug; it does not correspond to a guideline recommendation, which does not exist, and the series by Parodi and colleagues applies a more restrictive treatment threshold, limited to exudative neovascularisation [76]. A complementary paediatric case, of a 6-year-old boy with BVMD and a BCVA of 20/40 in the right eye and 20/160 in the left, treated with bevacizumab 1.25 mg/0.05 mL (two injections in the left eye and one in the right), ended at 9 months with 20/50 in the left eye, 20/30 in the right and scarred CNV, without adverse effects [104].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis by entity

In BVMD, visual deterioration is slow and visual acuity may vary over a lifetime between 20/20 and less than 20/200 [1]. Substantial visual loss is associated with the vitelliruptive and atrophic stages and with exudative CNV, and eyes with CNV have worse visual acuity [23, 56, 72]. Macular sensitivity on microperimetry showed no significant overall decline over a median of 4.4 years, but initially normal points lost 0.36 dB/year [73]. In AVMD, visual acuity lies between 20/40 and 20/200, with slow deterioration [1]. In ARB, visual acuity ranges from 20/20 to less than 20/200; the median age at severe visual impairment was 47 years with angle closure and 68 years without angle closure, with no significant annual linear change in visual acuity over a median follow-up of 3 years, and decline is slower in the absence of CNV and angle-closure glaucoma [1, 60, 72]. In ADVIRC, visual acuity remains good until late stages, when the whole retina is involved [1]. In the IMPG1/IMPG2 forms, mean visual acuity in the French series was 20/40, and the monoallelic IMPG2 form maintained a mean visual acuity of 0.15 and −0.01 logMAR, whereas the biallelic form progressed to 1.90 and 1.82 logMAR at 68.7 years [11, 24]. The sources consulted do not include longitudinal data on retention of a driving licence or on the mean age at legal blindness in BVMD. For driving in France, the PNDS sets a visual acuity limit of 5/10 and the absence of a central scotoma, and provides for periodic reassessment of fitness (section 11.2) [21].

11.2. Surveillance

GeneReviews recommends an annual ophthalmological examination, with BCVA, visual fields and SD-OCT, to monitor the lesions and detect CNV, and an annual examination in children to prevent amblyopia [1]. For the macular dystrophies, the AAO recommends follow-up every 1–2 years, with the examinations shown in the table [85]. The French PNDS recommends an annual examination by the treating ophthalmologist and review at a reference or competence centre every two to three years (the text states three years in the section on the frequency of visits), shortened in the event of loss of acuity or new difficulties and lengthened when acuity stabilises; each visit includes visual field, autofluorescence and SD-OCT, adjustment of the optical correction, a search for macular oedema and neovascularisation, and may include a psychology and genetics consultation and assessment of fitness to drive (visual acuity and binocular visual field) and for the workplace [21]. Patients should see the ophthalmologist in the event of reduced vision or metamorphopsia, and the Amsler grid may be used for self-assessment in some cases [1]. In practice, the annual surveillance of GeneReviews applies to children and to patients with active lesions, and the 1–2-year interval of the AAO to adults with stable disease.

Fundus photography should be used sparingly in the maculopathies, because of the potential for light toxicity, and short-wavelength autofluorescence with reduced illumination where possible [85]. Smoking cessation is part of follow-up [1].

12. Summary and Concluding Remarks

The bestrophinopathies form a spectrum of eye diseases caused by variants in BEST1, a gene that encodes a calcium-activated anion channel in the basolateral membrane of the RPE. BVMD, dominant and of childhood onset, is defined by a macular vitelliform lesion that progresses through stages, with an abnormal EOG and a normal full-field ERG; AVMD is the adult-onset form; ARB, recessive, is more extensive and is associated with hypermetropia and angle-closure glaucoma; ADVIRC, caused by variants that alter splicing, involves the whole globe. Adult-onset foveomacular vitelliform dystrophy is genetically heterogeneous, with PRPH2, IMPG1 and IMPG2 among the identified causes, and a normal EOG and the location of the material above the RPE point towards the IMPG1 and IMPG2 forms.

Diagnosis combines multimodal imaging, EOG according to the ISCEV standard and genetic testing by panel with CNV analysis, interpreted with attention to the number of alleles, phase and nomenclature on transcript NM_004183.4; some published notations, including that of the Swedish founder variant p.(Trp93Cys), do not correspond to the current reference sequence. Counselling distinguishes the 50% risk of the dominant forms, with incomplete penetrance and variable expressivity, from the 25% risk to sibs in ARB, whose carriers have a normal EOG.

There is no approved treatment and no disease-specific guideline for Best disease; the 2021 French PNDS covers it together with the genetic maculopathies, with annual surveillance, review at a reference centre every 2–3 years, low-vision rehabilitation and its own rules on driving and on testing minors. OCTA has shown that CNV is frequent in BVMD but mostly non-exudative and stable; intravitreal anti-VEGF is reserved for exudative or suspected exudative neovascularisation, with ranibizumab within the generic CNV indication in adults in the EU and the other uses off-label. In ARB, angle closure is the main determinant of severe visual impairment, gonioscopy should be part of the assessment and filtering surgery carries a risk of malignant glaucoma. OPGx-BEST1 gene therapy has preclinical canine efficacy and safety data and preliminary results from five patients released by the sponsor, without peer review; allele-specific editing and base editing remain at the preclinical stage. Ongoing natural history studies and the microperimetry endpoints proposed in 2026 are preparing the ground for controlled trials. Epidemiological data from Portugal and Brazil are not included in the sources consulted.

13. References

1. MacDonald IM, Lee T, Lawrence J. Bestrophinopathies. 2003 Sep 30 [updated 2020 Jul 16]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. PMID: 20301346. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1167/
2. Iannaccone A, Kerr NC, Kinnick TR, Calzada JI, Stone EM. Autosomal recessive best vitelliform macular dystrophy: report of a family and management of early-onset neovascular complications. Arch Ophthalmol. 2011;129(2):211-7. doi:10.1001/archophthalmol.2010.367. PMID: 21320969.
3. Burgess R, Millar ID, Leroy BP, Urquhart JE, Fearon IM, De Baere E, et al. Biallelic mutation of BEST1 causes a distinct retinopathy in humans. Am J Hum Genet. 2008;82(1):19-31. doi:10.1016/j.ajhg.2007.08.004. PMID: 18179881. PMC2253971.
4. Yardley J, Leroy BP, Hart-Holden N, Lafaut BA, Loeys B, Messiaen LM, et al. Mutations of VMD2 splicing regulators cause nanophthalmos and autosomal dominant vitreoretinochoroidopathy (ADVIRC). Invest Ophthalmol Vis Sci. 2004;45(10):3683-9. doi:10.1167/iovs.04-0550. PMID: 15452077.
5. Orphanet. Adult-onset foveomacular vitelliform dystrophy (ORPHA:99000). Available from: https://www.orpha.net/en/disease/detail/99000
6. GeneGuidance. IMPG1 — clinical gene curation. Export of 26 September 2026, accessed 10 October 2026.
7. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166. PMC5765404.
8. Krämer F, White K, Pauleikhoff D, Gehrig A, Passmore L, Rivera A, et al. Mutations in the VMD2 gene are associated with juvenile-onset vitelliform macular dystrophy (Best disease) and adult vitelliform macular dystrophy but not age-related macular degeneration. Eur J Hum Genet. 2000;8(4):286-92. doi:10.1038/sj.ejhg.5200447. PMID: 10854112.
9. Burgess R, MacLaren RE, Davidson AE, Urquhart JE, Holder GE, Robson AG, et al. ADVIRC is caused by distinct mutations in BEST1 that alter pre-mRNA splicing. J Med Genet. 2009;46(9):620-5. doi:10.1136/jmg.2008.059881. PMID: 18611979.
10. GeneGuidance. PRPH2 — clinical gene curation. Export of 26 September 2026, accessed 10 October 2026.
11. GeneGuidance. IMPG2 — clinical gene curation. Export of 26 September 2026, accessed 10 October 2026.
12. Manes G, Meunier I, Avila-Fernández A, Banfi S, Le Meur G, Zanlonghi X, et al. Mutations in IMPG1 cause vitelliform macular dystrophies. Am J Hum Genet. 2013;93(3):571-8. doi:10.1016/j.ajhg.2013.07.018. PMID: 23993198. PMC3769927.
13. Brandl C, Schulz HL, Charbel Issa P, Birtel J, Bergholz R, Lange C, et al. Mutations in the Genes for Interphotoreceptor Matrix Proteoglycans, IMPG1 and IMPG2, in Patients with Vitelliform Macular Lesions. Genes (Basel). 2017;8(7). doi:10.3390/genes8070170. PMID: 28644393. PMC5541303.
14. Orphanet. Best vitelliform macular dystrophy (ORPHA:1243). Available from: https://www.orpha.net/en/disease/detail/1243
15. Orphanet. Autosomal recessive bestrophinopathy (ORPHA:139455). Available from: https://www.orpha.net/en/disease/detail/139455
16. Orphanet. Autosomal dominant vitreoretinochoroidopathy (ORPHA:3086). Available from: https://www.orpha.net/en/disease/detail/3086
17. Fahim AT, Daiger SP, Weleber RG. Nonsyndromic Retinitis Pigmentosa Overview. 2000 Aug 4 [updated 2023 Apr 6]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1417/
18. Bitner H, Schatz P, Mizrahi-Meissonnier L, Sharon D, Rosenberg T. Frequency, genotype, and clinical spectrum of best vitelliform macular dystrophy: data from a national center in Denmark. Am J Ophthalmol. 2012;154(2):403-412.e4. doi:10.1016/j.ajo.2012.02.036. PMID: 22633354.
19. Dalvin LA, Pulido JS, Marmorstein AD. Vitelliform dystrophies: Prevalence in Olmsted County, Minnesota, United States. Ophthalmic Genet. 2017;38(2):143-147. doi:10.1080/13816810.2016.1175645. PMID: 27120116. PMC5638045.
20. Beryozkin A, Sher I, Ehrenberg M, Zur D, Newman H, Gradstein L, et al. Best Disease: Global Mutations Review, Genotype-Phenotype Correlation, and Prevalence Analysis in the Israeli Population. Invest Ophthalmol Vis Sci. 2024;65(2):39. doi:10.1167/iovs.65.2.39. PMID: 38411968. PMC10910552.
21. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Maculopathies génétiques. Centre de Référence Maladies Rares «Maladies Sensorielles Génétiques» (MAOLYA), Filière SENSGENE. Saint-Denis La Plaine: HAS; 2021 Jul 28. Available from: https://www.has-sante.fr/jcms/p_3280082
22. Boon CJ, van den Born LI, Visser L, Keunen JE, Bergen AA, Booij JC, et al. Autosomal recessive bestrophinopathy: differential diagnosis and treatment options. Ophthalmology. 2013;120(4):809-20. doi:10.1016/j.ophtha.2012.09.057. PMID: 23290749.
23. Haldrup SB, McClements ME, Cehajic-Kapetanovic J, Corydon TJ, MacLaren RE. Gene Therapy Strategies for the Treatment of Bestrophinopathies. Int J Mol Sci. 2025;26(19):9421. doi:10.3390/ijms26199421. PMID: 41096689. PMC12524932.
24. Meunier I, Manes G, Bocquet B, Marquette V, Baudoin C, Puech B, et al. Frequency and clinical pattern of vitelliform macular dystrophy caused by mutations of interphotoreceptor matrix IMPG1 and IMPG2 genes. Ophthalmology. 2014;121(12):2406-14. doi:10.1016/j.ophtha.2014.06.028. PMID: 25085631.
25. Constable PA, Bach M, Frishman LJ, Jeffrey BG, Robson AG. ISCEV Standard for clinical electro-oculography (2017 update). Doc Ophthalmol. 2017;134(1):1-9. doi:10.1007/s10633-017-9573-2. PMID: 28110380. PMC5309273.
26. European Medicines Agency. Lucentis (ranibizumab): EPAR — product information (Summary of Product Characteristics). EU authorisation 2007-01-22; latest EC decision 2026-09-28. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/lucentis
27. Genentech, Inc. LUCENTIS (ranibizumab injection), for intravitreal use. Prescribing information. BLA125156. Initial U.S. approval: 2006. Current revision 2025-11-21. DailyMed setid de4e66cc-ca05-4dc9-8262-e00e9b41c36d. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfoSetid.cfm?setid=de4e66cc-ca05-4dc9-8262-e00e9b41c36d
28. Opus Genetics, Inc. Safety and tolerability of subretinally injected OPGx-BEST1 in patients with Best vitelliform macular dystrophy (BVMD) or autosomal-recessive bestrophinopathy (ARB) (BIRD-1). ClinicalTrials.gov identifier: NCT07185256. Updated 2026 Sep 24 [cited 2026 Oct 10]. Available from: https://clinicaltrials.gov/study/NCT07185256
29. Crowley C, Paterson R, Lamey T, McLaren T, De Roach J, Chelva E, et al. Autosomal recessive bestrophinopathy associated with angle-closure glaucoma. Doc Ophthalmol. 2014;129(1):57-63. doi:10.1007/s10633-014-9444-z. PMID: 24859690. PMC4074353.
30. Sun Z, Liu C, Liu W, Zhang M, Ren S, Gao L, et al. Surgical treatment of autosomal recessive bestrophinopathy with angle-closure glaucoma: vitreous liquefaction as the key to correcting postoperative malignant glaucoma-three case reports. Front Med (Lausanne). 2025;12:1560475. doi:10.3389/fmed.2025.1560475. PMID: 41244779. PMC12615444.
31. Shmueli O, Batash T, Nitzan I, Chowers I, Tiosano L. Cataract surgery in eyes with adult-onset foveomacular-vitelliform dystrophy. Graefes Arch Clin Exp Ophthalmol. 2026;264(2):355-364. doi:10.1007/s00417-025-07018-x. PMID: 41182355. PMC12923482.
32. European Reference Network for Rare Eye Diseases (ERN-EYE). Home page and consensus statements [Internet]. [cited 2026 Oct 10]. Available from: https://www.ern-eye.eu/ ; https://www.ern-eye.eu/guidelines/
33. Best F. II. Über eine hereditäre Maculaaffektion. Z Augenheilkd. 1905;13:199-212. doi:10.1159/000290318.
34. Nordström S. Hereditary macular degeneration—a population survey in the county of Västerbotten, Sweden. Hereditas. 1974;78(1):41-62. doi:10.1111/j.1601-5223.1974.tb01427.x. PMID: 4448697.
35. Stöhr H, Marquardt A, Rivera A, Cooper PR, Nowak NJ, Shows TB, et al. A gene map of the Best's vitelliform macular dystrophy region in chromosome 11q12-q13.1. Genome Res. 1998;8(1):48-56. doi:10.1101/gr.8.1.48. PMID: 9445487. PMC310689.
36. Petrukhin K, Koisti MJ, Bakall B, Li W, Xie G, Marknell T, et al. Identification of the gene responsible for Best macular dystrophy. Nat Genet. 1998;19(3):241-7. doi:10.1038/915. PMID: 9662395.
37. Marquardt A, Stöhr H, Passmore LA, Krämer F, Rivera A, Weber BH. Mutations in a novel gene, VMD2, encoding a protein of unknown properties cause juvenile-onset vitelliform macular dystrophy (Best's disease). Hum Mol Genet. 1998;7(9):1517-25. doi:10.1093/hmg/7.9.1517. PMID: 9700209.
38. Sun H, Tsunenari T, Yau KW, Nathans J. The vitelliform macular dystrophy protein defines a new family of chloride channels. Proc Natl Acad Sci U S A. 2002;99(6):4008-13. doi:10.1073/pnas.052692999. PMID: 11904445. PMC122639.
39. Davidson AE, Millar ID, Urquhart JE, Burgess-Mullan R, Shweikh Y, Parry N, et al. Missense mutations in a retinal pigment epithelium protein, bestrophin-1, cause retinitis pigmentosa. Am J Hum Genet. 2009;85(5):581-92. doi:10.1016/j.ajhg.2009.09.015. PMID: 19853238. PMC2775838.
40. Kane Dickson V, Pedi L, Long SB. Structure and insights into the function of a Ca(2+)-activated Cl(-) channel. Nature. 2014;516(7530):213-8. doi:10.1038/nature13913. PMID: 25337878. PMC4454446.
41. Yang T, Liu Q, Kloss B, Bruni R, Kalathur RC, Guo Y, et al. Structure and selectivity in bestrophin ion channels. Science. 2014;346(6207):355-9. doi:10.1126/science.1259723. PMID: 25324390. PMC4341822.
42. Orphanet. MRCS syndrome (ORPHA:263347). Available from: https://www.orpha.net/en/disease/detail/263347
43. Felbor U, Schilling H, Weber BH. Adult vitelliform macular dystrophy is frequently associated with mutations in the peripherin/RDS gene. Hum Mutat. 1997;10(4):301-9. doi:10.1002/(SICI)1098-1004(1997)10:4<301::AID-HUMU6>3.0.CO;2-J. PMID: 9338584.
44. GeneGuidance. BEST1 — clinical gene curation. Export of 26 September 2026, accessed 10 October 2026.
45. Guziewicz KE, Cideciyan AV, Beltran WA, Komáromy AM, Dufour VL, Swider M, et al. BEST1 gene therapy corrects a diffuse retina-wide microdetachment modulated by light exposure. Proc Natl Acad Sci U S A. 2018;115(12):E2839-E2848. doi:10.1073/pnas.1720662115. PMID: 29507198. PMC5866594.
46. Zhao Q, Kong Y, Kittredge A, Li Y, Shen Y, Zhang Y, et al. Distinct expression requirements and rescue strategies for BEST1 loss- and gain-of-function mutations. Elife. 2021;10. doi:10.7554/eLife.67622. PMID: 34061021. PMC8169119.
47. Ni RL, Procopio R, Siebert E, Thompson JA, Lamey TM, McLaren TL, et al. Gain of function in BEST1: photoreceptor changes and myopia in autosomal dominant vitreoretinochoroidopathy. Ophthalmic Genet. 2026;47(4):380-391. doi:10.1080/13816810.2026.2635605. PMID: 41991505.
48. Ji C, Li Y, Kittredge A, Hopiavuori A, Ward N, Yao P, et al. Investigation and Restoration of BEST1 Activity in Patient-derived RPEs with Dominant Mutations. Sci Rep. 2019;9(1):19026. doi:10.1038/s41598-019-54892-7. PMID: 31836750. PMC6910965.
49. Eksandh L, Bakall B, Bauer B, Wadelius C, Andréasson S. Best's vitelliform macular dystrophy caused by a new mutation (Val89Ala) in the VMD2 gene. Ophthalmic Genet. 2001;22:107-15. PMID 11449320.
50. Mullins RF, Oh KT, Heffron E, Hageman GS, Stone EM. Late development of vitelliform lesions and flecks in a patient with best disease: clinicopathologic correlation. Arch Ophthalmol. 2005;123:1588-94. PMID 16286623.
51. Khan KN, Islam F, Moore AT, Michaelides M. The fundus phenotype associated with the p.Ala243Val BEST1 mutation. Retina. 2018;38:606-13. PMID 28225368.
52. Boon CJ, Klevering BJ, Leroy BP, Hoyng CB, Keunen JE, den Hollander AI. The spectrum of ocular phenotypes caused by mutations in the BEST1 gene. Prog Retin Eye Res. 2009;28(3):187-205. doi:10.1016/j.preteyeres.2009.04.002. PMID: 19375515.
53. Querques G, Zerbib J, Santacroce R, et al. The spectrum of subclinical Best vitelliform macular dystrophy in subjects with mutations in BEST1 gene. Invest Ophthalmol Vis Sci. 2011;52:4678-84. PMID 21436265.
54. Milenkovic A, Weber BHF. Allele-specific suppression of pathogenic bestrophin-1 transcripts by CRISPR/Cas9-mediated genome editing. Genome Med. 2026;18(1). doi:10.1186/s13073-026-01649-3. PMID: 42010707. PMC13104400.
55. Sæther E, Bertelsen G, Bragadottir R, Holtan JP. Clinical and genetic characterization of BEST1-associated retinal dystrophies in the Norwegian population. Acta Ophthalmol. 2026;104(6):682-690. doi:10.1111/aos.70065. PMID: 41627933. PMC13460395.
56. Han IC, Coussa RG, Mansoor M, Critser DB, Sohn EH, Russell JF, et al. Choroidal Neovascularization Is Common in Best Vitelliform Macular Dystrophy and Plays a Role in Vitelliform Lesion Evolution. Ophthalmol Retina. 2023;7(5):441-449. doi:10.1016/j.oret.2022.11.014. PMID: 36528270. PMC10164042.
57. Kalmodia S, Aparicio JG, Stepanian K, Beck A, Salas A, Harutyunyan N, et al. Allele-specific correction of dominant Best vitelliform macular dystrophy in patient-derived retinal pigment epithelium [preprint]. bioRxiv. 2026 Jul 15. doi:10.64898/2026.07.09.737567.
58. Bitner H, Mizrahi-Meissonnier L, Griefner G, Erdinest I, Sharon D, Banin E. A homozygous frameshift mutation in BEST1 causes the classical form of Best disease in an autosomal recessive mode. Invest Ophthalmol Vis Sci. 2011;52:5332-8. PMID 21467170.
59. Lu R, Zhou L, Tang L, Li X, Cao G. Case Report: Acute angle-closure glaucoma as the initial presentation of autosomal recessive bestrophinopathy caused by compound heterozygous BEST1 mutations. Front Med (Lausanne). 2026;13:1861665. doi:10.3389/fmed.2026.1861665. PMID: 42774095. PMC13593448.
60. Bianco L, Antropoli A, Boulanger-Scemama E, Nassisi M, Benadji A, Mésa G, et al. Autosomal Recessive Bestrophinopathy-Phenotypic Variability, Natural History, and Genotype-Phenotype Correlations. Am J Ophthalmol. 2026;283:188-198. doi:10.1016/j.ajo.2025.12.012. PMID: 41421761.
61. Formenti F, Salvetti AP, Boon CJF, Invernizzi A, Staurenghi G, Romano F. Circinate Midperipheral Vitelliform Dystrophy With Macular Sparing in the Father of a Patient With Best Disease. Retina. 2026;46(1):e3-e5. doi:10.1097/IAE.0000000000004578. PMID: 40591937.
62. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 304 (Retinal receptor dystrophies).
63. MacDonald IM, Gudiseva HV, Villanueva A, Greve M, Caruso R, Ayyagari R. Phenotype and genotype of patients with autosomal recessive bestrophinopathy. Ophthalmic Genet. 2012;33:123-9. PMID 21809908.
64. Wivestad Jansson R, Berland S, Bredrup C, Austeng D, Andréasson S, Wittström E. Biallelic mutations in the BEST1 gene: additional families with autosomal recessive bestrophinopathy. Ophthalmic Genet. 2016;37:183-93. PMID 26333019.
65. Zhao L, Grob S, Corey R, et al. A novel compound heterozygous mutation in the BEST1 gene causes autosomal recessive Best vitelliform macular dystrophy. Eye (Lond). 2012;26:866-71. PMID 22422030.
66. Johnson AA, Guziewicz KE, Lee CJ, et al. Bestrophin 1 and retinal disease. Prog Retin Eye Res. 2017;58:45-69. PMID 28153808.
67. Michaelides M, Urquhart J, Holder GE, Restori M, Kayali N, Manson FD, et al. Evidence of genetic heterogeneity in MRCS (microcornea, rod-cone dystrophy, cataract, and posterior staphyloma) syndrome. Am J Ophthalmol. 2006;141(2):418-20. doi:10.1016/j.ajo.2005.09.018. PMID: 16458719.
68. Weleber RG, Carr RE, Murphey WH, Sheffield VC, Stone EM. Phenotypic variation including retinitis pigmentosa, pattern dystrophy, and fundus flavimaculatus in a single family with a deletion of codon 153 or 154 of the peripherin/RDS gene. Arch Ophthalmol. 1993;111(11):1531-42. PMID: 8240110.
69. Boon CJ, den Hollander AI, Hoyng CB, Cremers FP, Klevering BJ, Keunen JE. The spectrum of retinal dystrophies caused by mutations in the peripherin/RDS gene. Prog Retin Eye Res. 2008;27(2):213-35. PMID: 18328765.
70. Olivier G, Corton M, Intartaglia D, Verbakel SK, Sergouniotis PI, Le Meur G, et al. Pathogenic variants in IMPG1 cause autosomal dominant and autosomal recessive retinitis pigmentosa. J Med Genet. 2021;58(8):570-578. PMID: 32817297.
71. Al-Khuzaei S, Shalaby AK, Yu J, Shanks M, Clouston P, MacLaren RE, et al. Expanding the genetic spectrum in IMPG1 and IMPG2 retinopathy. Genes (Basel). 2025;16(12):1474. doi:10.3390/genes16121474. PMID: 41465146.
72. Shi J, Tian L, Sun TY, Xu K, Xie Y, Zhang XH, et al. Bestrophinopathies: Clinical Characteristics, Natural History, and Genetic Landscape. Ophthalmol Retina. 2026;10(4):452-464. doi:10.1016/j.oret.2025.10.014. PMID: 42559932.
73. Bianco L, Forte G, Antropoli A, Del Fabbro S, Doddato G, Calzavara S, et al. Natural History of Macular Sensitivity in Best Vitelliform Macular Dystrophy: Microperimetry-Derived Outcome Measures in Preparation For Clinical Trials. Invest Ophthalmol Vis Sci. 2026;67(11):1. doi:10.1167/iovs.67.11.1. PMID: 42678224. PMC13549402.
74. Polosa A, Lu M, Dorfman AL, Masis-Solano M, Costantino S, Qian CX. Characterization of functional and structural impairments in best vitelliform macular dystrophy using visual electrophysiology and optical coherence tomography in pediatric and adult patients. Doc Ophthalmol. 2026;152(2):129-144. doi:10.1007/s10633-025-10064-9. PMID: 41222609.
75. Battaglia Parodi M, Mansour A, Saladino A, Arrigo A, Cicinelli MV, Vilela M, et al. ULTRA-WIDEFIELD EXAMINATION OF THE PERIPHERAL RETINA IN BEST1 -RELATED RETINOPATHY. Retina. 2026;46(8):1424-1430. doi:10.1097/IAE.0000000000004846. PMID: 41914891.
76. Parodi MB, Arrigo A, Bandello F. Optical Coherence Tomography Angiography Quantitative Assessment of Macular Neovascularization in Best Vitelliform Macular Dystrophy. Invest Ophthalmol Vis Sci. 2020;61(6):61. doi:10.1167/iovs.61.6.61. PMID: 32602906. PMC7415901.
77. Battaglia Parodi M, Bianco L, Antropoli A, Nunziata A, Del Fabbro S, Arrigo A, et al. Acquired Torpedo-like Maculopathy Associated With Macular Neovascularization in Best Vitelliform Macular Dystrophy. Retina. 2026;46(4):e28-e29. doi:10.1097/IAE.0000000000004666. PMID: 40924902.
78. Voide N, Eandi C, Gaillard MC, Kaeser PF, Hoeckele N, Munier F, et al. Choroidal Neovascularisation Secondary to Best Vitelliform Macular Dystrophy: Paediatric Cases Monitored by OCT and OCT-A. Klin Monbl Augenheilkd. 2026;243(4):475-478. doi:10.1055/a-2810-4233. PMID: 42008983.
79. Kritfuangfoo T, Pandit A, Feo A, Faghihi S, Sarraf D. Best vitelliform macular dystrophy masquerading as pachyvitelliform maculopathy in an elderly patient. Can J Ophthalmol. 2026. doi:10.1016/j.jcjo.2026.08.008. PMID: 42632626.
80. Besalti Ekinci Z, Ozbek M, Artunay O. Central retinal vein occlusion in BEST1-related angle-closure glaucoma in autosomal recessive bestrophinopathy: a case report. Ophthalmic Genet. 2026:1-4. doi:10.1080/13816810.2026.2658716. PMID: 41987504.
81. Liu J, Xue Y, Huang W, Wang Y. Micropulse transscleral cyclophotocoagulation in the treatment of autosomal recessive bestrophinopathy combined with angle closure glaucoma: a case report. Front Med (Lausanne). 2025;12:1567964. doi:10.3389/fmed.2025.1567964. PMID: 40861203. PMC12370765.
82. Testa F, Rossi S, Passerini I, et al. A normal electro-oculography in a family affected by Best disease with a novel spontaneous mutation of the BEST1 gene. Br J Ophthalmol. 2008;92:1467-70. PMID 18703557.
83. Padhy SK, Habjan MŠ, Constable PA. Review of the clinical electrooculogram - Part 2: the bestrophinopathies and modified protocols. Doc Ophthalmol. 2026;152(3):287-307. doi:10.1007/s10633-026-10093-y. PMID: 41795754. PMC13194349.
84. Padhy SK, Constable PA. The short wavelength electro-oculogram (SW-EOG) in best disease: preliminary results. Doc Ophthalmol. 2026;152(2):117-127. doi:10.1007/s10633-026-10081-2. PMID: 41629643. PMC13083315.
85. Duncan JL, Branham K, Birch DG, Daiger SP, Ferris FL 3rd, Fishman GA, et al. Guidelines on clinical assessment of patients with inherited retinal degenerations [Clinical Statement]. San Francisco (CA): American Academy of Ophthalmology; 2022. Available from: https://aao.org/Assets/fb574a6a-a13a-4353-bcbd-34d767a7147d/638006550690630000/inherited-retinal-degenerations-clinical-statement-2022-pdf
86. Kinnick TR, Mullins RF, Dev S, et al. Autosomal recessive vitelliform macular dystrophy in a large cohort of vitelliform macular dystrophy patients. Retina. 2011;31:581-95. PMID 21273940.
87. Orphanet. Pattern dystrophy (ORPHA:63454). Available from: https://www.orpha.net/en/disease/detail/63454
88. Orphanet. Butterfly-shaped pigment dystrophy (ORPHA:99001). Available from: https://www.orpha.net/en/disease/detail/99001
89. Pawloff M, Hollaus M, Mylonas G, Pircher M, Hitzenberger CK, Pfeifer M, et al. Multimodal Imaging of Dual BEST1/EFEMP1-Associated Hereditary Macular Disease. J Clin Med. 2026;15(14). doi:10.3390/jcm15145495. PMID: 42513409. PMC13412455.
90. Antropoli A, Bianco L, Zanlonghi X, Benadji A, Condroyer C, Antonio A, et al. Phenotypic and Genotypic Characterization of RP1L1-Associated Retinopathy. Invest Ophthalmol Vis Sci. 2025;66(4):7. doi:10.1167/iovs.66.4.7. PMID: 40172514. PMC11967996.
91. Mahler EA, Moeller LC, Wall K, Saßmannshausen M, Kron B, Bolz HJ, et al. Mutation of the Thyroid Hormone Receptor Beta Gene (THRB) Causes Vitelliform Macular Dystrophy with High Intrafamilial Variability. Genes (Basel). 2025;16(10). doi:10.3390/genes16101240. PMID: 41153457. PMC12564572.
92. Rajavoor Muniswamy JK, Stewart M, Eggenberger E, Muthusamy K. Vitelliform retinal lesions: Uncommon adult-onset presentation of the common mitochondrial MELAS variant. Mol Genet Metab. 2026;149(1-2):110245. doi:10.1016/j.ymgme.2026.110245. PMID: 42648229.
93. Orphanet. Acute exudative polymorphous vitelliform maculopathy (ORPHA:715855). Available from: https://www.orpha.net/en/disease/detail/715855
94. Opus Genetics, Inc. Opus Genetics announces positive low dose cohort 1 data from phase 1/2 clinical trial of OPGx-BEST1 and successful FDA Type C meeting with potential phase 3 dosing in 2027 [press release; Exhibit 99.1, Form 8-K]. Research Triangle Park (NC); 2026 Sep 9. Available from: https://www.sec.gov/Archives/edgar/data/1228627/000114036126035979/ef20081753_ex99-1.htm
95. Chaudhary KM, Mititelu M, Lieberman RM. An evidence-based review of vascular endothelial growth factor inhibition in pediatric retinal diseases: part 2. Coats' disease, best disease, and uveitis with childhood neovascularization. J Pediatr Ophthalmol Strabismus. 2013;50(1):11-9. doi:10.3928/01913913-20120821-02. PMID: 22938515.
96. Marano F, Deutman AF, Leys A, Aandekerk AL. Hereditary retinal dystrophies and choroidal neovascularization. Graefes Arch Clin Exp Ophthalmol. 2000;238:760-4. PMID 11045344.
97. Leu J, Schrage NF, Degenring RF. Choroidal neovascularisation secondary to Best's disease in a 13-year-old boy treated by intravitreal bevacizumab. Graefes Arch Clin Exp Ophthalmol. 2007;245:1723-5. PMID 17605026.
98. Querques G, Zerbib J, Santacroce R, et al. Functional and clinical data of Best vitelliform macular dystrophy patients with mutations in the BEST1 gene. Mol Vis. 2009;15:2960-72. PMID 20057903.
99. Andrade RE, Farah ME, Costa RA. Photodynamic therapy with verteporfin for subfoveal choroidal neovascularization in best disease. Am J Ophthalmol. 2003;136:1179-81. PMID 14644242.
100. Sodi A, Murro V, Caporossi O, Passerini I, Bacci GM, Caputo R, et al. Long-Term Results of Photodynamic Therapy for Choroidal Neovascularization in Pediatric Patients with Best Vitelliform Macular Dystrophy. Ophthalmic Genet. 2015;36(2):168-74. doi:10.3109/13816810.2015.1009121. PMID: 25675349.
101. Regeneron Pharmaceuticals, Inc. EYLEA (aflibercept) injection, for intravitreal use. Prescribing information. BLA125387. Initial U.S. approval: 2011. Current revision 2026-07-07. DailyMed setid f96cfd69-da34-41ee-90a9-610a4655cd1c. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfoSetid.cfm?setid=f96cfd69-da34-41ee-90a9-610a4655cd1c
102. European Medicines Agency. Eylea (aflibercept): EPAR — therapeutic indications. EU authorisation 2012-11-21; latest EC decision 2026-01-15. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/eylea
103. European Medicines Agency. Lytenava (bevacizumab gamma): EPAR — therapeutic indications. EU authorisation 2024-05-27. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/lytenava
104. Chhablani J, Jalali S. Intravitreal bevacizumab for choroidal neovascularization secondary to Best vitelliform macular dystrophy in a 6-year-old child. Eur J Ophthalmol. 2012;22(4):677-9. doi:10.5301/ejo.5000095. PMID: 22139615.
105. Opus Genetics, Inc. Form 10-Q, quarterly report for the period ended June 30, 2026. Washington (DC): U.S. Securities and Exchange Commission; 2026 Aug 6. Available from: https://www.sec.gov/Archives/edgar/data/1228627/000162828026054381/ird-20260630.htm
106. Gray A, Kwok JC, Sato Y, Assenmacher CA, Choudhary M, Jayagopal A, et al. Evaluation of the Toxicity and Efficacy of an Adeno-Associated Viral Vector Expressing BEST1 Delivered by Subretinal Injection in a Canine Model of Human Bestrophinopathy. Hum Gene Ther. 2026;37(17-18):876-886. doi:10.1177/10430342261458076. PMID: 42411064. PMC13401768.
107. Sinha D, Steyer B, Shahi PK, Mueller KP, Valiauga R, Edwards KL, et al. Human iPSC Modeling Reveals Mutation-Specific Responses to Gene Therapy in a Genotypically Diverse Dominant Maculopathy. Am J Hum Genet. 2020;107(2):278-292. doi:10.1016/j.ajhg.2020.06.011. PMID: 32707085. PMC7413860.
108. Foundation Fighting Blindness. Opus Genetics launches gene therapy clinical trial for Best disease. 2025 Nov 14. Available from: https://www.fightingblindness.org/news/opus-genetics-launches-gene-therapy-clinical-trial-for-best-disease-3148
109. Phase 1/2 clinical trial of OPGx-BEST1 shows positive results for inherited retinal diseases. Ophthalmology Times. 2026 Sep 10. Available from: https://www.ophthalmologytimes.com/view/phase-1-2-clinical-trial-of-opgx-best1-shows-positive-results-for-inherited-retinal-diseases
110. Opus Genetics reports positive phase 1/2 data for OPGx-BEST1 in BEST1-related retinal disease. Eyewire+. 2026 Sep 9. Available from: https://eyewire.news/news/opus-genetics-reports-positive-phase-12-data-for-opgx-best1-in-best1-related-retinal-disease
111. Columbia University. Study of BEST1 vitelliform macular dystrophy (natural history). ClinicalTrials.gov identifier: NCT05809635. Updated 2025 Jul 30 [cited 2026 Oct 10]. Available from: https://clinicaltrials.gov/study/NCT05809635
112. Clemons TE, Milton RC, Klein R, Seddon JM, Ferris FL 3rd. Risk factors for the incidence of advanced age-related macular degeneration in the Age-Related Eye Disease Study (AREDS) AREDS report no. 19. Ophthalmology. 2005;112:533-9. PMID 15808240.
113. Leenders M, Gaastra M, Tuller S, Magrath G, Jayagopal A, Malone KE. Review and meta-analysis of BEST1 retinopathy: global prevalence and mutational landscape [preprint]. medRxiv. 2026 Jan 16. doi:10.64898/2026.01.15.26343613.