Choroideremia

Progressive X-linked chorioretinal dystrophy caused by loss of function of Rab escort protein 1 (REP-1, encoded by the CHM gene), which begins with night blindness in childhood, progresses to centripetal loss of the visual field and preserves central vision until the fourth or fifth decade, with manifestations of variable severity in heterozygous women.

Literature and regulatory information updated: 31 October 2026.

0. Initial Reference Table

1. Introduction

1.1 Clinical description and nomenclature

Choroideremia is an X-linked inherited chorioretinal degeneration that affects the choriocapillaris, the retinal pigment epithelium (RPE) and the photoreceptors, and whose characteristic lesion is scalloped chorioretinal atrophy of the mid-periphery with macular sparing [1]. The first description is attributed to Mauthner in 1872, an indirect attribution made by the authors of a later case report [12]. The name, formed by analogy with irideremia (absence of the iris), would mean «absence of the choroid»; OMIM notes that the term is a misnomer, because there is no congenital absence of the choroid and the disease is an abiotrophy that begins shortly after birth and progresses slowly, which is why Waardenburg preferred the term «tapetochoroidal dystrophy» (TCD) [1]. In 1934 Wilmer documented a 35-year-old man with «choroidal sclerosis» whose disease had been transmitted through unaffected women over several generations, and Sorsby's descriptions of «X-linked choroidal sclerosis» came to be interpreted as cases of choroideremia; in 1974 Shapiro and Gorlin concluded that choroidal sclerosis is a stage in the evolution of the same disease [1].

Orphanet defines choroideremia (ORPHA:180) as an X-linked chorioretinal dystrophy with progressive degeneration of the choroid, RPE and retina, with onset in childhood, adolescence or adulthood, and classifies it among rare genetic diseases and rare ophthalmological diseases [2]. The Oxford Desk Reference describes a diffuse, progressive condition that begins with night blindness in childhood, evolves to peripheral field loss and, usually by the age of 50 years, to loss of central vision; in the early stages the fundus may be confused with retinitis pigmentosa (RP), although choroideremia is not primarily a photoreceptor dystrophy, and in the late stages the appearance of extensive atrophy of the choriocapillaris and RPE is characteristic [19].

1.2 Milestones of the molecular discovery

The locus was mapped to Xq13–q21 by linkage studies in the 1980s (lod score of 5.78 with the marker DXYS1, with no recombination), and the genealogical study of the Finnish families showed that 36 patients and 48 carriers shared the same haplotype, consistent with a single founder allele [1]. Visible interstitial deletions in Xq21, some in men with choroideremia associated with intellectual disability and deafness, and balanced X-autosome translocations in affected women made it possible to narrow the region [1, 3]. In 1990 Cremers et al. cloned the gene by reverse genetics strategies: they isolated eight overlapping cDNA clones from Xq21, expressed in the retina, choroid and RPE, whose reading frame was disrupted in eight patients with deletions and in a woman with a balanced translocation involving Xq21 [20]. In 1994 van Bokhoven et al. characterised the complete reading frame, with 15 exons and a protein of 653 amino acids [5].

The function of the protein was clarified by Seabra, Brown and Goldstein: the choroideremia gene product resembles component A of Rab geranylgeranyl transferase, and patients' lymphoblasts have a marked deficiency of component A activity, but not of component B; the deficiency was more pronounced with Rab3A than with Rab1A, which implied the existence of more than one component A protein [21]. This functional redundancy explains why loss of one of the isoforms causes degeneration in the organ in which that isoform is most needed [1]. Saudubray includes choroideremia, in the table of retinopathies of disorders of cellular trafficking, in the vesicular trafficking category [22].

1.3 From characterisation to gene therapy

The size of the CHM cDNA (about 1.96 kb), compatible with adeno-associated virus (AAV) vectors, the slow course and the prolonged preservation of the macula made choroideremia a model for retinal gene therapy [23]. The first phase 1/2 trial of subretinal AAV2-REP1 was published in 2014 [24], and the phase 3 STAR trial, published in 2023, did not meet its primary endpoint [9]. In 2026 there is no approved disease-modifying treatment [13, 25].

1.4 Scope of the article

This article covers CHM-associated choroideremia in hemizygous men and heterozygous women, including mosaic forms, and the Xq21 contiguous gene deletion syndrome (OMIM 303110), in which choroideremia is associated with deafness due to loss of POU3F4 and with intellectual disability [3, 4]. The general approach to inherited retinal dystrophies and X-linked RP, gyrate atrophy of the choroid and retina, Usher syndrome and POU3F4-associated deafness are part of the differential diagnosis.

2. Molecular Genetics and Pathophysiology

2.1 The CHM gene and the REP-1 protein

The CHM gene lies at Xq21.2, has 15 exons and encodes REP-1, a cytosolic protein of 653 amino acids expressed in all cells [4, 5]. The MANE Select reference transcript is NM_000390.4, which encodes NP_000381.1 [6]. REP-1 is the substrate-binding component A of Rab geranylgeranyl transferase: it binds unprenylated Rab GTPases and presents them to the catalytic dimer formed by RABGGTA and RABGGTB, which adds 20-carbon geranylgeranyl groups to their terminal cysteine residues [5, 7]. The Rab proteins regulated by this lipid modification control endocytosis, exocytosis and intracellular vesicular trafficking [4]. Crystal structures of REP-1 with Rab7 show that Rab interacts with a binding platform of REP-1 through the switch 1 and 2 regions, and that the carboxyl terminus of REP-1 acts as a mobile lid over a hydrophobic region that coordinates the Rab termini [5].

The existence of more than one protein with component A function, shown by the residual activity in patients' lymphoblasts, creates functional redundancy: loss of one of the isoforms causes degeneration in the tissue in which that isoform is most needed, which is consistent with a disease limited to the eye [1, 21]. CHM is a gene on the long arm of the X that is expressed from the inactive X in cell lines, at levels that vary between individuals, which places it among the genes that partially escape inactivation [5].

2.2 Cellular and tissue mechanism

In the absence of REP-1, Rab proteins are underprenylated, and degeneration affects the choroid, the RPE and the photoreceptors [7]. Accumulation of unprenylated Rab proteins is associated with abnormal phagocytosis of photoreceptor outer segments and reduced melanosome transport, leading to RPE death [26]. The RPE processes are the most affected, but the photoreceptors are also involved [4]. In the conditional Chm knockout mouse, photoreceptor and RPE degeneration are triggered independently and are associated with different subsets of defective Rab proteins; histopathological study of the eye of a symptomatic 88-year-old carrier also showed apparently independent loss of photoreceptors and RPE, with a normal choriocapillaris except where the retina was severely degenerated, and localised REP-1 to the cytoplasm of the rods [1]. Constitutive Chm knockout in the mouse is lethal in male embryos, because REP-1 is essential in the extraembryonic membranes [5].

2.3 Inheritance pattern

Choroideremia is transmitted in an X-linked manner, with no male-to-male transmission [4]. Hemizygous men have the full disease; heterozygous women are cellular mosaics as a result of X inactivation, which begins around the ninth day after fertilisation, and almost always have fundus signs, although most are asymptomatic [4, 27]. The sources differ in the label: Orphanet classifies it as X-linked recessive, GeneReviews and the OMIM table as X-linked, and the OMIM genemap2 file as X-linked dominant [1, 2, 4]. The divergence reflects the frequent expression in heterozygotes and does not alter recurrence risks.

Skewed X inactivation explains some of the heterozygotes with a severe picture, but not all: skewing greater than 80:20 occurs in about 9% of women without disease (≥90:10 in 3%, ≥70:30 in 30%) [27], and in a Mexican series of 12 carriers inactivation was random in 11, including the severely affected carrier, with no consistent correlation between the inactivation pattern and the phenotype [1]. A woman with a male-type picture should raise suspicion of skewed inactivation, monosomy X (Turner syndrome), an X-autosome translocation disrupting the gene or somatic mosaicism [1, 27, 28].

2.4 Variant spectrum and nomenclature

Almost all known pathogenic variants are nonsense, small deletions or insertions, or splicing variants, and predict truncation or absence of REP-1; the mechanism is loss of function [4]. Sequencing detects up to 75% of probands' variants and deletion/duplication analysis about 25%, proportions of detectable variants by method and not clinical sensitivity [4]. The 2014 European clinical utility gene card counted more than 130 unique variants in the locus-specific LOVD database (Leiden Open Variation Database), with complete or partial gene deletions in 25–50% and nonsense variants in about 30%; deletions range from a few kilobases, removing one exon, to about 15 Mb [29]. In an analysis of 57 families, variants were found in 54: more than 42% were single-nucleotide substitutions, nearly 4% complete gene deletions, nearly 4% deletions/insertions, more than 9% large intragenic or partial deletions, about 28% deletions of fewer than 5 base pairs and about 13% splicing variants; recurrences clustered at C>T transitions in CGA arginine codons, which create a stop codon, and four of the five CGA codons in the gene were recurrence sites [5].

Alongside these classes there are mechanisms that escape routine testing: a complete insertion of an L1 retrotransposon into exon 6, which causes direct splicing from exon 5 to exon 7 while maintaining the reading frame; an SVA retrotransposon insertion detectable as a drop in coverage over exon 2 on next-generation sequencing; deep intronic variants creating cryptic exons; promoter variants and large duplications [4, 5]. A synonymous variant, c.1359C>T p.(Ser453=), identified in a Brazilian patient, causes aberrant splicing with loss of exon 11, predicting p.(Gln451PhefsTer3), demonstrated in peripheral blood RNA [30]. In the Finnish cohort of Lähteenoja et al., structural and copy-number variants accounted for 7% (5/70) of the variants in retinal dystrophies, and the first pathogenic intragenic inversion of CHM was described [8]. The five most frequent CHM variants account for only 11% of the alleles described, which limits gene editing directed at specific variants [23].

Missense variants are rare: c.1520A>G p.(His507Arg), described in a 21-year-old Italian man, prevents the interaction of REP-1 with Rab geranylgeranyl transferase and excludes the protein from the isoprenylation cycle, and NM_000390.4:c.1649T>C p.(Leu550Pro) (published as c.1679T>C), predicted to destabilise the structure, was associated with absence of REP-1 in lymphocytes [5, 29, 31]. A novel missense variant has low predictive value until there are biochemical studies of the mutant protein [29]. In the older nomenclature, variants were numbered by exons designated by letters (B3, B4, C) or by an earlier reference sequence, which requires each historical designation to be confirmed in a normalisation tool [5].

2.5 Founder variants

A 13-generation family from the Salla region, in north-eastern Finnish Lapland, which in the 1990s represented about one fifth of the patients with choroideremia known worldwide, shares the intron 13 splicing variant now described as c.1609+2dup; this variant does not cause the disease in the other four Finnish families studied [5, 31]. Lymphoblast RNA from these patients showed two aberrantly spliced transcripts and no normal transcript [5].

2.6 Genotype-phenotype correlations, penetrance and expressivity

GeneReviews does not recognise any demonstrated genotype-phenotype correlations [4]. Nevertheless, splicing variants that leave very low levels of correctly processed mRNA may give milder phenotypes: according to the authors of the STAR trial, 2–5% of normal transcript may significantly attenuate progression [9]. The primary data come from an Oxford series: two unrelated men with the variant NM_000390.4:c.940+3del (described as c.940+3delA), outside the canonical donor sequence, had large areas of residual autofluorescence for their age and longer degeneration half-lives than a cohort of 30 patients; they expressed full-length CHM transcript at 2.3% and 4.7% of the predominant truncated transcript, equivalent to less than about 1% of the level in unaffected people, with REP-1 below the immunoblot detection threshold, whereas a patient with c.940+2T>A at the same donor site, with typical progression, had no detectable full-length transcript [6, 32]. Complete gene deletions are associated with severe disease at an early age: in one family the proband had at 4 years a hypopigmented fundus, RPE mottling and reduced scotopic ERG, and at 6 years severe atrophy of the RPE and choriocapillaris [1]. In hemizygous men the positive predictive value of a pathogenic variant is estimated at more than 99%, and non-penetrance has not been described [29]. Expressivity varies even within the same family: in the Salla family some men were almost blind before the age of 30 and others were asymptomatic after 50 [1].

2.7 Xq21 contiguous gene deletion and other allelic forms

Large interstitial deletions of Xq21 cause choroideremia with severe intellectual disability and malformations such as cleft lip and palate and agenesis of the corpus callosum; smaller deletions cause choroideremia, mixed sensorineural and conductive deafness due to loss of POU3F4 and intellectual disability of variable degree attributed to loss of RPS6KA6 (RSK4) [4]. The deletions described measure between about 6 and 16 Mb; some patients have short stature, hypotonia, motor delay or ataxia, and the deafness is accompanied by stapes fixation and a perilymph gusher at stapedectomy, as in DFNX2 deafness (OMIM 304400) [3]. Computed tomography of the temporal bones shows a defective cochlear modiolus with communication between the basal turn of the cochlea and the internal auditory canal, a finding considered pathognomonic of POU3F4-associated deafness [3, 33]. A 3-year-old Korean child with a 16 Mb deletion did not yet have signs of choroideremia, which is explained by the progressive nature of the dystrophy; the mother, a carrier of the deletion, had mild high-frequency hearing loss [3]. A woman with a de novo X;4 translocation disrupting Xq21 had premature ovarian insufficiency and mixed deafness [4]. The X-autosome translocations disrupting CHM, t(X;7)(q21.2;p12), t(X;13)(q21.2;p12) and t(X;4)(q21.2;p16.3), have been described only in women, with mild signs of choroideremia and ovarian dysgenesis, and a complex t(X;1;3) translocation was found in a carrier with a severe phenotype and ectodermal dysplasia [29]. Isolated DFNX2 deafness is caused by variants of POU3F4, not of CHM.

3. Clinical Spectrum and Manifestations by System

3.1 Time course in hemizygous men

The disease almost always begins before adolescence with difficulty seeing in the dark, progresses to peripheral field loss and only late affects central vision [4]. In a retrospective study of 71 men cited by the authors of the STAR trial, the mean age at onset of night blindness was 12.6 years (±1.0) and that of peripheral visual loss 19.7 years (±1.3); central vision remains good until about the age of 40 years and then declines rapidly once degeneration reaches the fovea, in a phase that may last 5 to 10 years until acuity is no longer measurable [9]. Men in the fifth decade frequently have very good acuity with a small visual field, and by around 50 to 70 years central vision is markedly impaired [4]. Cross-sectional studies place legal blindness between the fifth and seventh decades [26].

3.2 Fundus and imaging

The changes begin as patchy areas of chorioretinal degeneration in the mid-periphery, which progress to marked loss of the RPE and choriocapillaris with preservation of the large deep choroidal vessels; peripapillary RPE atrophy is early and progressive, and the function and anatomy of the central macula are preserved until late, as shown by fundus autofluorescence [4]. In a series of 29 children from a London centre (mean age at referral 9 years, range 3–16), the first sign of retinopathy was diffuse pigment clumping at the level of the RPE, which progressed to more marked chorioretinal atrophy in the mid-periphery; peripapillary atrophy was early and progressive, autofluorescence showed three distinct zones, spectral-domain optical coherence tomography (SD-OCT) showed persistence of the inner retinal layers in 15/27 children, and subfoveal choroidal thickness decreased with age while central retinal thickness increased [11]. At the last visit, at a mean age of 16 years (7–26), most maintained excellent acuity (0.98 ± 0.13 on the decimal scale), and CHM variants were identified in 15 of the 19 tested [11]. In some Asian series the fundus is atypical: 6 of 157 Chinese probands (4%) with a diagnosis of RP had CHM variants identified by exome sequencing, without the typical scalloped atrophy [1].

The area of preserved fundus autofluorescence (PFAF) and the width of the ellipsoid zone on SD-OCT are the structural measures of progression. A meta-analysis of longitudinal studies cited by Birch et al. estimated an exponential decline in residual RPE area of 0.05 log mm² per year (95% CI 0.046–0.055), corresponding to a loss of 10.88% per year [26].

3.3 Visual function

The ERG initially shows a pattern of rod-cone degeneration, with early loss of the rod response to dim scotopic stimuli in boys, followed by cone deterioration, and then becomes markedly reduced or non-recordable, with variability within and between families [29]; the visual field shows a ring scotoma whose areas of loss correspond to the areas of degeneration [4]. Colour vision is affected early: in a prospective study of 30 patients and 10 controls, the total error score on the Farnsworth-Munsell 100 hue test was 120 in patients with acuity better than 6/6, 206 in patients with worse acuity and 47 in controls, with a defect generalised across the spectrum that worsens when degeneration reaches the fovea [1].

3.4 Quantitative natural history: prospective cohorts

The meta-analysis by Shen et al. pooled 1004 eyes of 693 patients from 23 studies (3–84 years): best-corrected visual acuity (BCVA), measured in letters on the ETDRS (Early Treatment Diabetic Retinopathy Study) scale, follows a biphasic decline, with a transition at 39.1 years (95% CI 33.5–44.7), a loss of 0.24 letters per year (0.04–0.45) before and of 1.16 letters per year (0.93–1.39) after; in the longitudinal data (102 eyes, 287 visits), the decline before 39 years (0.33 letters/year) did not differ from zero, and afterwards it was 1.23 letters/year (p=0.004) [10]. The correlation between the two eyes was moderate (r=0.60), and the median age at reaching low vision, subnormal vision and legal blindness (≤35 letters) was 45, 52 and 66 years in the worse eye and 47, 58 and 72 years in the better eye [10].

The prospective, multicentre NIGHT study followed for 20 months men aged 18 years or older with a genetic diagnosis, active disease at the macula and BCVA of at least 34 letters; 220 completed the study, mean BCVA remained stable, and 81.4% (worse eye) and 77.8% (better eye) lost fewer than 5 letters; at 12 months, retinal sensitivity decreased by 0.68 and 0.48 dB, the central ellipsoid zone area by 0.276 and 0.290 mm² and the total autofluorescence area by 0.605 and 0.533 mm², with low symmetry between eyes [34]. The mean age of NIGHT participants was 47.1 years [26].

The natural history study by Birch et al. (5 centres in the USA and Canada, recruitment between December 2016 and June 2018) included 57 men aged 14 to 70 years (mean 32.4 ± 11.2), with BCVA of at least 20/200 and a preserved paracentral ellipsoid zone, assessed every six months for up to 48 months; 47 of 55 completed 24 months and 26 of 55 completed 48 months, mainly because participants left to join gene therapy trials [26]. Genotypes were deletions in 43.9% and point variants in 40.4%. Baseline BCVA was 81.9 (±8.9) and 81.1 (±9.5) letters, and the annual loss −0.17 and −0.16 letters (not significant), equivalent to about one letter in six years; at 48 months, 92% (right eye) and 77% (left eye) of eyes had changed by fewer than 5 letters, and none had lost 15 or more [26].

The correlation between autofluorescence area and the V30 index of static perimetry was high (R² of 0.76 and 0.59), and the authors concluded that the area of preserved autofluorescence is the most suitable primary endpoint for trials, because BCVA is insensitive to progression before foveal involvement [26]. There were no serious adverse events related to the examinations, and two participants underwent bilateral cataract surgery during follow-up [26]. In advanced disease, partial collapse of the fovea («foveal splitting») makes acuity readings variable [9]. A retrospective series of 115 patients, cited by Palejwala et al., found a mean age of 39 years, 79% with acuity of at least 20/40, severe loss (≤20/200) mainly after the age of 60 and a loss of about one line in 5 years [12].

3.5 Ocular complications

Posterior subcapsular cataract is similar to that of RP; cystoid macular oedema has been described; retinal detachment is more frequent with high myopia; and choroidal neovascularisation is rare [4]. In the AAO IRIS registry (Intelligent Research in Sight), in the USA, between 2013 and 2019, 540 patients with a new coded diagnosis of choroideremia (mean age 57.2 years) were compared with 38,179 patients with RP: in choroideremia, cataract was recorded in 40.9%, cataract surgery in 10.4%, rhegmatogenous detachment in 5.6%, its repair in 1.3%, cystoid macular oedema in 12.6% and choroidal neovascularisation in 6%; the risk of detachment (hazard ratio, HR, 1.73; 95% CI 1.18–2.54) and of repair (HR 3.37; 1.57–7.22) was higher than in RP, with no significant difference for cataract and macular oedema [35]. The data depend on coding with ICD-10-CM code H31.21 and may include diagnostic errors [35].

In an SD-OCT series cited by Palejwala et al., cystoid macular oedema was present in 10 of 16 patients (63%) [12]. Choroidal neovascularisation tends to arise in the second or third decade in eyes with preserved submacular choriocapillaris, is small, poorly exudative and may regress; by 2014 five cases had been described, at 14, 17, 27 (carrier), 46 (carrier) and 72 years [12]. In 2026 a full-thickness macular hole was described in a 30-year-old man, with syneretic vitreous and a thin, friable internal limiting membrane [36], a retinal detachment associated with progressive peripheral retinoschisis in a 43-year-old man [37] and extensive macular schisis treated by vitrectomy [38].

3.6 Heterozygous women

Most heterozygotes are asymptomatic, but have signs of depigmentation and RPE atrophy visible on careful examination, more evident after the second decade; night blindness and field loss may appear later through expansion of the atrophic areas, and mild symptomatic forms are probably under-reported [4]. The ERG is normal or reduced (6 of 7 normal and 1 with reduced amplitudes in one series), the fundus shows pigment stippling and chorioretinal atrophy, the desaturated D15 test may reveal colour deficits, the fields range from normal to defects similar to those of men, and autofluorescence shows patchy areas of loss of fluorescence throughout the fundus [4]. Carrier autofluorescence has a speckled pattern that distinguishes them from carriers of X-linked RP due to RP2 or RPGR, in whom the pattern is radial [1].

In Finland only 7 of 105 carriers reported symptoms, but 21 of 52 examined had field changes and abnormal dark adaptation, more marked in older women [1]. In a series of 12 carriers from Columbia University, the spectrum ranged from mild (RPE mottling, hypoautofluorescent stippling, intact retinal layers) to severe, similar to the male phenotype (extensive atrophy and central islands with scalloped borders), with 4 severe cases; of the 4 with follow-up, 2 lost acuity [39]. In Oxford, 17 carriers (34 eyes; median age 44 years, 15–73; median follow-up 7 years, 1–11) were classified into fine (5 eyes), coarse (13), geographic (12) and male-pattern (4) phenotypes; those with a severe phenotype had significant loss of retinal sensitivity (−0.7 and −0.8 dB/year) and expansion of atrophy, and the authors recommend not relying on the classification alone [40]. In a Chinese series of 64 carriers, a new classification based on autofluorescence and fundus photography distinguished granular (76.3%), severe peripapillary atrophy (7.5%), localised atrophy (10.8%) and diffuse atrophy (5.4%) types; the granular type remained stable, the others progressed (mean atrophy expansion of 3.1 mm²/year), neither age nor genotype was associated with the severe phenotype, and the baseline phenotype was the best predictor (agreement κ=0.93) [41].

A 24-year-old Japanese woman with a male-pattern phenotype (diffuse chorioretinal atrophy, a star-shaped macular island of preserved autofluorescence and an ERG non-recordable under scotopic and photopic conditions, with preserved acuity) had the frameshift variant c.525_526del with an allele fraction of 10% in leucocytes, not detected by Sanger sequencing in blood but present in the heterozygous state in buccal mucosa, consistent with somatic mosaicism; X inactivation in buccal mucosa was skewed (83.1:16.9), with phase relative to the variant not determined [28]. In the Finnish cohort, symptomatic women with choroideremia attributed to skewed X inactivation were identified [8].

3.7 Extraocular manifestations

Choroideremia associated with intragenic CHM variants has no systemic manifestations described in the sources consulted. Deafness, intellectual disability, malformations and premature ovarian insufficiency point to a contiguous Xq21 deletion or a chromosomal rearrangement (section 2.7) [3, 4]. In Kearns-Sayre syndrome some patients have a retinopathy with a choroideremia pattern and complete choroidal atrophy, a mitochondrial phenocopy [42].

4. Clinical Recognition and Warning Signs

4.1 When to suspect

Choroideremia should be considered in a boy or man with difficulty adapting to the dark, a fundus with patchy mid-peripheral chorioretinal atrophy and peripapillary atrophy, a compatible ERG and fields and a family history consistent with X-linked inheritance, without male-to-male transmission; the absence of a family history does not exclude the diagnosis [4]. In the child, the earliest sign is diffuse pigment clumping in the RPE with peripapillary atrophy, at a stage when acuity is normal [7, 11]. In men with a previous diagnosis of RP and a not very typical fundus, choroideremia should be considered, because it accounts for a fraction of RP cases without a molecular diagnosis in some populations [1, 9].

4.2 Warning signs and high-value combinations

4.3 Newborn screening

The sources consulted do not describe newborn screening for choroideremia or biochemical biomarkers applicable for that purpose; presymptomatic diagnosis is made by targeted testing in known families (section 8).

5. Diagnosis: Criteria, Tests and Algorithms

5.1 Diagnostic criteria

There are no formal clinical criteria published by a society. GeneReviews defines suspicion in a man with difficulty adapting to the dark, a characteristic fundus, compatible ERG and visual fields and a family history consistent with X-linked inheritance, and considers the diagnosis established in a man with these findings and a hemizygous pathogenic variant in CHM; a hemizygous variant of uncertain significance (VUS) neither establishes nor excludes the diagnosis [4]. In a woman, heterozygosity is diagnosed by identification of the familial variant or, when no affected man is available, by sequencing followed by deletion/duplication analysis [4].

5.2 Ophthalmological assessment at the initial visit

The 2022 AAO recommendations for the assessment of inherited retinal degenerations group them into four classes and include CHM-associated choroideremia, together with gyrate atrophy, in class 3 (chorioretinal degenerations) [15]. For this class, the initial visit includes ocular and medical history (with retinotoxic drugs) and family history, pedigree, BCVA by the ETDRS (Early Treatment Diabetic Retinopathy Study) protocol, colour vision (optional), slit-lamp biomicroscopy, intraocular pressure, indirect ophthalmoscopy, colour or wide-field fundus photography, SD-OCT, short-wavelength autofluorescence with reduced illumination, near-infrared reflectance or autofluorescence, kinetic and static perimetry, microperimetry, full-field ERG when appropriate and the full-field stimulus threshold (FST) test when fixation is unstable or the ERG is non-recordable [15]. Static perimetry and microperimetry are of uncertain value in advanced disease with eccentric, unstable fixation, and an ERG that is no longer detectable need not be repeated [15]. GeneReviews adds assessment of macular structure by SD-OCT (ellipsoid zone, outer retinal tubulations, extent of atrophy and associated maculopathy) [4].

5.3 Molecular tests

GeneReviews proposes two approaches: with a suggestive phenotype, single-gene testing (sequencing first and, if negative, targeted deletion/duplication analysis) or a multigene retinal dystrophy panel; when the diagnosis has not been considered, exome or genome sequencing [4]. The AAO recommends genetic testing for most patients with a presumed genetic retinal degeneration, accepts single-gene testing, a panel or an exome at the initial visit, generally reserves the genome for research and allows the analysis to be repeated when the first result was negative or inconclusive, because gene discovery and variant interpretation evolve [15]. The 2022 document refers to the 2012 AAO task force recommendations on genetic testing of inherited eye diseases [15, 44]. The Oxford Desk Reference states that, with a good phenotype and family history, retinal dystrophy panels give about a 55% probability of a genetic diagnosis across the dystrophies as a whole, and that panels of 100 to 150 genes identify the cause in about 90% of X-linked RP [19, 45]. In the northern Finland cohort, 22 of 51 patients with choroideremia (43%) had been tested, with a molecular diagnosis in 20 (91%) [8].

The proportions in the table are proportions of variants detectable by method among probands with an established diagnosis, not the clinical sensitivity of the test. The European clinical utility gene card estimates the analytical sensitivity and specificity of Sanger sequencing at more than 98%, recommends MLPA for deletions and duplications, particularly in heterozygotes, and reports that, in a Moorfields cohort, 2 of 120 families with a typical phenotype and linkage to CHM had no variant identified; as choroideremia is not genetically heterogeneous, clinical sensitivity in large families with a typical phenotype equals analytical sensitivity, but phenocopies apparently reduce it in small families [29]. When the familial variant is known, the negative predictive value of a targeted test approaches 100% [29]. When the phenotype is suggestive and DNA testing is negative, GeneReviews recommends studying CHM expression by RNA and REP-1 by immunoblot; if REP-1 is absent, intronic and promoter variants, retrotransposon insertions and large duplications are sought, which may be revealed incidentally by a polymerase chain reaction (PCR) product larger than expected or by a drop in coverage on next-generation sequencing [4]. In 1998 MacDonald et al. proposed immunoblotting with an anti-REP-1 antibody in peripheral blood as a simple confirmation of the diagnosis [1].

5.4 Mosaicism and heterozygotes

Non-molecular detection of heterozygotes is most reliable with autofluorescence, which shows patchy loss at any age; fundus examination is only reliable after the age of 25, and a normal fundus before that age does not exclude heterozygosity [4, 7]. No non-molecular test is definitive in a man [7]. In boys, fundus changes usually appear before the age of 5, which gives the examination high clinical specificity; in young girls, retinal examination and ERG cannot exclude heterozygosity, which requires genetic testing [29]. Somatic mosaicism in a woman may escape Sanger sequencing in blood: in the 2026 Japanese case, the variant had an allele fraction of 10% in leucocytes on exome sequencing and was only confirmed in the heterozygous state in buccal mucosa [28]. In women with severe atrophy and no family history, testing should include CHM and, if necessary, a second tissue [28].

5.5 Report content and interpretation

The report should state the gene, the transcript with version (NM_000390.4), the HGVS nomenclature at DNA and protein level, the zygosity (hemizygous in the man, heterozygous in the woman), the classification according to the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria and the method that detected the variant, including copy-number analysis. GeneReviews notes that the notations in its variant table follow the original authors without verification; the historical designations in the variant table in section 2.4 are normalised in VariantValidator or ClinVar [4, 6, 31]. A hemizygous VUS may be reclassified through family segregation (with autofluorescence of the women in the maternal line), RNA analysis and REP-1 immunoblot [1, 4].

5.6 Diagnostic algorithm

5.7 Prenatal and preimplantation diagnosis

When the familial pathogenic variant is known, prenatal testing and preimplantation genetic diagnosis are possible [4]. The Oxford Desk Reference mentions prenatal diagnosis with molecular confirmation [19]. Reproductive guidance is developed in section 8.

6. Differential Diagnosis

6.1 Comparative table

6.2 Main differentials

RP is the most frequent misdiagnosis, because the early fundus may be confused with RP and because the end stages of the two diseases converge; the absence of intraretinal bone spicules, early peripapillary atrophy, visible deep choroidal vessels and the carrier autofluorescence pattern point towards choroideremia [1, 4, 19]. The distinction from X-linked RP has practical consequences, because gene therapy trials are gene-specific. Gyrate atrophy produces scalloped chorioretinal atrophy of a highly specific appearance, affects both sexes with autosomal recessive inheritance and is distinguished by raised plasma ornithine; a subgroup responds to pyridoxine and an arginine-restricted diet is the mainstay of treatment [19, 43]. In Usher syndrome type 1, profound congenital deafness and vestibular areflexia separate it from isolated choroideremia, but the combination of choroideremia and deafness should prompt a search for the contiguous Xq21 deletion, in which the deafness is mixed, with stapes fixation [3, 4]. The mitochondrial phenocopy of Kearns-Sayre syndrome is recognised by ophthalmoplegia, ptosis and cardiac conduction abnormalities [42].

7. Treatment and Risk Management Strategies

7.1 Principles and available recommendations

As of 31 October 2026, no disease-modifying treatment for choroideremia had been approved by the US Food and Drug Administration (FDA) or by the European Commission, and the European and US orphan designations for vectors carrying the CHM gene have not led to marketing authorisations [13, 25, 46]. GeneReviews states that no clinical practice guidelines specific to choroideremia have been published [4]. The reference for assessment and follow-up is the 2022 AAO document on inherited retinal degenerations, which places choroideremia in class 3 [15]. No ERN-EYE consensus document specific to choroideremia was identified.

Management rests on five axes: baseline assessment and multimodal follow-up; treatment of complications (cataract, cystoid macular oedema, retinal detachment, choroidal neovascularisation, macular hole); protection and nutrition; low-vision rehabilitation and psychological, educational and occupational support; and access to clinical trials and registries [4, 15, 19].

7.2 Protection, nutrition and general measures

GeneReviews recommends ultraviolet (UV) filtering spectacles outdoors, particularly in the face of light reflected from water and snow, and avoidance of tobacco, which is a major risk factor for macular degeneration [4]. It recommends an antioxidant supplement only for those without access to fresh fruit and green leafy vegetables, finds no information on the efficacy of vitamin A in choroideremia, and considers a regular source of very-long-chain omega-3 fatty acids, such as docosahexaenoic acid (DHA) from fish, possibly beneficial [4]. These recommendations are not supported by trials in choroideremia: a simvastatin trial (NCT01654562) was terminated with 2 patients because of insufficient recruitment, and the DARC study of vitamin A palmitate (NCT05045703) was withdrawn before recruiting [47].

7.3 Treatment of ocular complications

Posterior subcapsular cataract is treated surgically, and retinal detachment by conventional surgery [4]. In the IRIS registry, detachment repair was more than three times as frequent as in RP (HR 3.37; 95% CI 1.57–7.22) [35]. Vitreoretinal surgery in these eyes has particular features described in 2026 reports: in a 30-year-old man with a full-thickness macular hole, vitrectomy with internal limiting membrane peeling found syneretic vitreous and a thin, friable membrane, gas reabsorption was delayed, and the hole was closed at 6 months with a good visual outcome [36]; in a 43-year-old man with the c.242_245del variant and a detachment associated with peripheral retinoschisis that progressed to the fovea (acuity from 20/20 to 20/100), high posterior scleral buckling with cryotherapy led to recovery to 20/25 [37]; and vitrectomy with internal limiting membrane peeling has been described for extensive macular schisis [38]. Cystoid macular oedema is monitored by SD-OCT; Palejwala et al. cite a small case series with promising results from topical dorzolamide, with no controlled trial in choroideremia [4, 12]. In that series, two men aged 45 and 63 years with cystoid macular oedema on SD-OCT were treated with dorzolamide 2% eye drops, one drop three times a day in each eye; macular thickness decreased in both eyes at 2 months and further at 5 months, ETDRS acuity improved in at least one eye of each patient, and contrast sensitivity and microperimetry did not change in a clinically significant way [48]. This use is off-label: the dorzolamide label (Trusopt® and generics) approves it for raised intraocular pressure in ocular hypertension and open-angle glaucoma, with the same dosing of one drop in the affected eye three times a day, and paediatric safety and efficacy were demonstrated in a 3-month trial in that indication [49].

7.4 Intravitreal anti-VEGF for choroidal neovascularisation

The rare choroidal neovascularisation of choroideremia has been treated with intravitreal bevacizumab, an off-label use [4, 12]. Bevacizumab is a humanised monoclonal anti-VEGF (vascular endothelial growth factor) antibody. There is no established dose for choroideremia; the dose used in the report by Palejwala et al. was 1.25 mg per intravitreal injection (Avastin® formulation), and the dose approved for an analogous condition, neovascular age-related macular degeneration, is the same [12, 50].

Extrapolation of the dose from macular degeneration to choroideremia is stated as such: no trial has studied bevacizumab in this disease, and the paediatric age of the single report falls outside the indication of the SmPC [12, 50].

Ranibizumab, a humanised monoclonal antibody fragment directed against VEGF, has in the European Union an indication that covers choroidal neovascularisation of any cause in adults: the SmPC for Lucentis® (initial authorisation of 22/01/2007) includes, alongside neovascular AMD, diabetic macular oedema, proliferative diabetic retinopathy and macular oedema secondary to retinal vein occlusion, the «treatment of visual impairment due to choroidal neovascularisation» [14]. This indication rests on the MINERVA trial, randomised and double-masked, in which 178 adults with CNV from causes other than AMD and pathological myopia received ranibizumab 0.5 mg or sham injection (2:1); at 2 months, acuity changed by +9.5 letters against −0.4 letters, with an effect in every aetiological subgroup (in the subgroup of uncommon aetiologies, with 37 patients, the effect against sham was 10.6 letters), and the mean was 5.8 injections over 12 months [14]. Dosing in adults is 0.5 mg (0.05 ml of the 10 mg/ml solution) per intravitreal injection, with a minimum interval of 4 weeks between injections in the same eye; in CNV, the number of injections is individualised according to disease activity, and some patients need only one injection in the first 12 months [14]. In adolescents aged 12 to 17 years, five patients treated with 0.5 mg in MINERVA gained 5 to 38 letters (mean 16.6) with a mean of 3 injections, but the SmPC makes no dosing recommendation below 18 years outside retinopathy of prematurity [14]. In the USA, the Lucentis® label restricts CNV to the myopic form (0.5 mg monthly initially), and the CNV of choroideremia falls outside the indication [52]. Choroidal neovascularisation of choroideremia in an adult therefore falls within the European and Brazilian indication for ranibizumab, although no patient with choroideremia is identified in MINERVA; in an adolescent, as in the case of Palejwala et al., any anti-VEGF is off-label use.

The most serious adverse effects of intravitreal ranibizumab relate to the injection (endophthalmitis, rhegmatogenous detachment, retinal tear and iatrogenic traumatic cataract), and intraocular inflammation and raised intraocular pressure require monitoring in the following week [14, 53]. In Brazil, the ANVISA open data (updated on 13/10/2026) record bevacizumab only in the antineoplastic class, and intravitreal use of the oncology formulations is off-label [55].

7.5 Visual rehabilitation and psychosocial support

The AAO recommends low-vision assessment at the initial visit and every 1–2 years as needed, liaison with the school, orientation and mobility training, vocational rehabilitation, information on guide dogs, and mental health support from diagnosis and throughout progression, including support groups and individual counselling [15]. GeneReviews includes in counselling depression, loss of independence, fitness to drive and employment-related anxiety [4]. Visual loss has a measurable effect on quality of life: in the study by Birch et al., the composite score of the VFQ-25 questionnaire (National Eye Institute Visual Function Questionnaire) fell by 4.2 points over 24 months and by 7.0 points over 48 months [26].

7.6 Gene replacement therapy: rationale and technique

Gene replacement therapy aims to restore REP-1 function and Rab prenylation in surviving RPE cells, halting RPE death and the secondary degeneration of the photoreceptors and the choroid [23]. The most studied vector, timrepigene emparvovec (BIIB111, AAV2-REP1), is an AAV2 carrying the normal CHM cDNA under the CAG promoter with an inactivated post-transcriptional regulatory element (WPRE), administered by a single subretinal injection of up to 100 µl after vitrectomy and creation of a bleb with saline solution [9]. The two-step technique (a saline bleb followed by injection of the medicinal product) is of particular importance in choroideremia, because the atrophy makes controlled retinal detachment difficult; the fovea is vulnerable to iatrogenic detachment, and intraoperative optical coherence tomography is recommended [23, 56]. The intravitreal route is limited by the internal limiting membrane and by greater immunogenicity, and the suprachoroidal route is poorly suited when the choroid is thin or absent [23].

7.7 Gene therapy clinical trials

The first phase 1/2 trial, in Oxford (NCT01461213), treated 6 men aged 35 to 63 years with subfoveal AAV.REP1 (0.6–1.0×10^10 genome particles); despite iatrogenic foveal detachment, two patients with advanced disease gained 21 and 11 letters, the four with near-normal acuity recovered to within 1 to 3 letters of baseline, the mean gain was 3.8 letters (standard error, SE, 4.1) and maximal sensitivity on dark-adapted microperimetry rose from 23.0 to 25.3 dB at 6 months, with a correlation between gain and dose per mm² of surviving retina (r=0.82) [24]. At 3.5 years, the gains of the two patients with advanced disease were maintained (+21 and +18 letters in the treated eye, against −18 and −6 letters in the control eyes), three of the four patients with good baseline acuity retained it, and the patient who lost acuity in both eyes (−29 letters in the treated eye and −18 in the control) had received a total vector dose lower than planned [57]. The final report of the Oxford trial, with 14 patients (13 treated with 1×10^10 or 1×10^11 genome particles), met the primary endpoint: at 2 years, median acuity in treated eyes improved by 4.5 letters (interquartile range −2.0 to 8.8) and that of untreated eyes fell by 1.5 letters (−4.8 to 0.0; p=0.04), six treated eyes gained more than 5 letters, and the gains were maintained to 5 years in the patients recruited first; there were two significant adverse events linked to administration, a retinal stretch with underdosing and retinal inflammation at 2 weeks probably related to the vector [58].

Trials with the same vector followed (1×10^11 genome particles in 0.1 ml, subretinal) in Alberta (NCT02077361), Miami (phase 2, NCT02553135) and Tübingen (THOR, NCT02671539), all completed [47]. In Alberta, 6 men aged 30 to 42 years were followed for 2 years: one lost 8 letters, another gained 15 or more, a similar improvement occurred in the untreated eye of a third, the area of preserved RPE decreased at the same rate in treated and untreated eyes, and one patient had a serious adverse event, a localised intraretinal immune reaction with marked loss of function and of the outer layers on SD-OCT [59]. In Miami, 6 men aged 32 to 72 years with advanced disease were treated in the worse eye: at 2 years, one treated eye improved by 10 letters and another by 5, one untreated eye improved by 4, and the rest were within 2 letters of baseline; there were no serious adverse events, two patients developed an atrophic retinal hole in a non-functioning macular area already thinned before surgery, and intraoperative optical coherence tomography made it possible to avoid excessive foveal stretching [60]. In THOR, an open-label randomised phase 2 trial with 6 men (mean age 54.9 years), the change in acuity at 24 months was +3.7 letters in treated eyes and 0.0 in control eyes (difference 3.7; 95% CI −7.2 to 14.5; p=0.43), with no difference in retinal sensitivity, and the 28 adverse events were consistent with vitrectomy, none serious; patients received oral prednisone 1.0 mg/kg from the day before until day 19 and moxifloxacin 0.5% and dexamethasone 0.5% eye drops four times a day for 21 days [61]. Across the phase 1/2 trials, 9% of eyes gained 15 or more letters at 24 months, with a mean gain of 5.6 letters in patients with moderate loss (34–73 letters) [9].

The phase 1/2 trial by Spark Therapeutics (AAV2-hCHM, NCT02341807) treated 15 patients aged 20 to 57 years with a low dose (up to 5×10^10 vector genomes, n=5) or a high dose (up to 1×10^11, n=10) subfoveally; there was no systemic or vector-related toxicity, acuity remained within 15 letters of baseline in 13 of the 15 at 2 years, one patient had acute foveal thinning and another a macular hole, and there were no significant differences in sensitivity or in the extent of preserved RPE [56]. The REGENERATE trial, funded by the UK National Institute for Health and Care Research (NIHR) (NCT02407678; 2 hospitals), treated 30 men aged 18 years or older with BCVA of at least 6/60 with up to 1×10^11 particles; the change in BCVA at 24 months was −2.63 letters (standard error 2.76) in the treated eye versus +2.67 (0.768) in the control eye (p=0.08), field loss was greater in treated eyes, possibly through a surgical effect, and 6 serious adverse events occurred in 4 patients; the authors found no evidence of efficacy [62].

The phase 3 STAR trial (NCT03496012; EudraCT 2015-003958-41) was randomised, with parallel groups and a masked assessor, and allocated 169 men aged 18 years or older, with a genetic diagnosis, active disease at the macula and BCVA between 34 and 73 letters (≤6/12 and ≥6/60) to a high dose (1.0×10^11 vector genomes, n=69), a low dose (1.0×10^10, n=34) or untreated control (n=66), in a 2:1:2 ratio, with no sham procedure for ethical reasons; all treated participants received oral corticosteroid for 21 days, starting 2 days before [9]. Recruitment ran from 11/12/2017 to 04/10/2019 at 17 centres; the mean age was 48.7 years, 87.8% of patients were white and mean baseline BCVA was 60.0 letters [9]. The primary endpoint, the proportion gaining at least 15 letters at 12 months, was not met: 3/65 (5%) in the high dose, 1/34 (3%) in the low dose and 0/62 in the control group (p=0.245 and p=0.354) [9]. In exploratory analyses, the mean difference in BCVA versus control was +2.1 letters (95% CI −2.0 to 6.2) in the high dose and +0.9 in the low dose; 13.8%, 17.6% and 1.6% gained 10 or more letters; and 83.1%, 70.6% and 67.7% maintained acuity or lost fewer than 5 letters [9]. The area of preserved autofluorescence decreased more in treated eyes (−0.1264 mm² versus control in the high dose) [9]. According to a 2025 review by the Oxford investigators, the 2-line gain was statistically significant as a secondary endpoint (p=0.017), and a 2-line gain would be acceptable to the European Medicines Agency (EMA), whereas the FDA had set the threshold at 3 lines [23].

Treatment-emergent adverse events occurred in 91% (high dose), 94% (low dose) and 51% (control), and ocular inflammation in 51%, 47% and 2%; there was one case of severe non-infectious retinitis, reductions in acuity were the most frequent serious ocular events (3 related to the drug and 7 to the procedure), cataract occurred in 14%, 12% and 5%, and there were no deaths and no withdrawals for adverse events [9]. The authors cite as limitations the advanced disease of the participants, an unrealistic 3-line endpoint and the COVID-19 pandemic, which required alternative assessments in 19% of patients; two participants were followed locally in Brazil because of travel restrictions [9]. In June 2021 Biogen announced that STAR had not met its primary endpoint and halted development of BIIB111 [63].

The GEMINI study (NCT03507686), an open-label phase 2 study, assessed sequential bilateral administration of timrepigene emparvovec 1×10^11 vector genomes (up to 0.1 ml) 12 months apart: 66 patients were treated and 53 completed; acuity was broadly maintained in both eyes, serious surgical complications occurred in 7.6%, retinal inflammation events in 45.5% (with similar rates in both eyes) and severe non-infectious retinitis in 2 patients (3.0%); previous exposure did not increase the risk in the second eye, there was no systemic immune reaction, and baseline neutralising antibodies were possibly associated with more inflammation and loss of acuity after the first eye [64]. The long-term follow-up study SOLSTICE (NCT03584165; 277 participants treated with BIIB111 or BIIB112) is registered as completed, with an update in July 2026, with no results published in the sources consulted [47, 63].

4D Molecular Therapeutics developed 4D-110, with the 4D-R100 capsid and a codon-optimised CHM for intravitreal injection (NCT04483440, phase 1 dose escalation, 13 participants, active not recruiting); Roche ended the collaboration in 2021 and the company ended development in January 2025 [47, 65, 66]. Other gene-independent approaches are in trial in RP and choroideremia: intravitreal optogenetics RTx-015 (Ray Therapeutics; ENVISION, NCT06460844, phase 1, about 18 patients, 5-year follow-up, active not recruiting) and the intravitreal molecular photoswitch KIO-301 (Kiora; ABACUS, NCT05282953, phase 1/2, completed); the published phase 1 trial of KIO-301 included patients with advanced RP, with no results specific to choroideremia [47, 67]. Retinal prostheses (Bionic Vision, IRIS) have included choroideremia among their indications [47].

The systematic review and meta-analysis by Chen et al. (search up to 03/12/2025) estimated, in addition to the gains in acuity and sensitivity, less RPE degeneration (mean difference −4.41; 95% CI −6.39 to −2.44) and an increase in subfoveal choroidal thickness (+9.13 µm; 7.53–10.72); pooled treatment-emergent adverse events were around 35%, mostly mild to moderate and procedure-related, and serious events were below 20%, with severe non-infectious retinitis in 3.0% [25]. In the individual data, 5 patients given 10^10 particles gained 4.4 ± 4.2 letters at 2 years versus −2.4 ± 1.9 in control eyes, and 8.4 ± 4.7 versus −8.8 ± 3.1 at 3.5 years; foveal thinning in 4 of 6 treated eyes, macular hole, localised intraretinal immune reaction and corticosteroid-induced ocular hypertension in 2 patients were described [25]. The authors interpret the effect as modest, stabilising and not restorative, acknowledge publication bias for RPE area and adverse events, and recall that one untreated eye gained 15 or more letters in one of the studies, which limits fellow-eye controls [25].

These results do not allow indirect comparisons between trials, because they differ in population (age, baseline acuity, area of preserved RPE), dose, vector, surgical technique and endpoint. The meta-analysis by Shen et al. showed that BCVA scarcely changes before the age of 39 and that symmetry between eyes is only moderate (r=0.60), which makes BCVA an insensitive endpoint in young patients and fellow-eye controls imperfect [10]. Natural history studies propose the area of preserved autofluorescence, the ellipsoid zone and retinal sensitivity as more sensitive endpoints [26, 34], and the Oxford investigators propose the «smooth zone» of autofluorescence, corresponding to still-healthy RPE, as a structural biomarker for early disease [23].

7.8 Regulatory status

An orphan designation, European or FDA, is not a marketing authorisation and does not demonstrate quality, safety or efficacy [13, 46]. The FDA orphan designation database records three designations for choroideremia, all with the status «designated» and none with approval [46]. The absence of authorisation applications in the public databases consulted does not exclude confidential applications.

7.9 Developments over the past 12 months (November 2025 to October 2026)

Preclinical research in 2025–2026 also includes iPSC-derived RPE models with CHM knockout, in which digestion of outer-segment phagosomes is impaired, with a doubling of undigested autofluorescent material and delayed acquisition of LAMP1, pointing to subtle lysosomal dysfunction and premature ageing of the RPE [69]. The low frequency of each variant (the five most frequent account for 11% of alleles) favours variant-independent approaches, such as gene replacement or mRNA delivery [23, 68].

7.10 Children, transition and special populations

In children, symptoms appear before adolescence, acuity is usually normal and structural changes (pigment clumping, peripapillary atrophy, choroidal thinning) progress while central function is maintained [4, 11]. The AAO allows adapted examination regimens or examination under sedation in young children, who often cannot perform the functional tests used in adults [15]. The Oxford Desk Reference recommends psychological and educational support and a multidisciplinary team [19]. At the transition to adulthood, the AAO recommends repeating genetic counselling when the disease began in childhood [15]. Gene therapy trials included adults only [25].

Heterozygotes with a severe phenotype should be followed like affected men, because they lose retinal sensitivity and RPE area over time, and the authors of the carrier series propose their inclusion in trials [39–41].

7.11 Pregnancy, perioperative care and agents to avoid

The sources consulted do not describe effects of pregnancy on the progression of choroideremia or specific precautions. In vitreoretinal surgery, the fragility of the internal limiting membrane, the syneretic vitreous and the delayed gas resorption described in a 2026 report justify dedicated planning [36]. UV exposure reflected from water and snow and smoking should be avoided [4]. The AAO includes a search for retinotoxic drugs in the history [15]. In the contiguous Xq21 deletion, stapedectomy caused a perilymph gusher in one patient, and CT of the temporal bones shows the cochlear malformation characteristic of POU3F4-associated deafness [3, 33].

8. Reproductive Options and Family Counselling

8.1 Recurrence risks

It is not possible to predict, in an affected boy, the age of onset or the rate of progression, nor to predict whether a heterozygote will have visual loss [4]. Mosaicism of early origin is documented: an L1 retrotransposition occurring early in embryonic development explained an isolated case [4]. In X-linked diseases with zero reproductive fitness, two thirds of the mothers of isolated cases are carriers; choroideremia does not reduce reproductive fitness in the same way, and that calculation does not apply directly, but the principle of offering testing to sisters even when the mother has a negative result, because of germline mosaicism, still holds [4, 27].

8.2 Assessment of the mother and cascade testing

The mother of an isolated case should be assessed by molecular testing and by fundus examination with dilated pupils; a normal fundus before the age of 25 may not exclude heterozygosity [4]. The Oxford Desk Reference recommends examining relatives, especially the mother of affected boys, and notes that the fundus and electrophysiology help in detecting carriers of choroideremia and X-linked RP, but that genetic testing should be used when available [19]. Identification of heterozygotes requires the familial variant to be known or, when no affected man is available, sequencing followed by deletion/duplication analysis [4]. Autofluorescence is the most reliable non-molecular means of identifying heterozygotes at any age [7]. Molecular testing can identify the family member in whom the variant arose de novo, which clarifies the risk for the extended family [4].

The AAO recalls that testing of asymptomatic relatives in the absence of established treatment should be carefully considered and accompanied by genetic counselling [15]. In minors at risk, the timing of testing is weighed with the family in the light of these implications; in girls, the result has mainly future reproductive implications.

8.3 Reproductive planning, prenatal and preimplantation diagnosis

The appropriate time to determine genetic risk and discuss prenatal and preimplantation diagnosis is before pregnancy, and counselling should be offered to affected, heterozygous or at-risk young adults [4]. With the familial pathogenic variant identified, prenatal testing in an at-risk pregnancy and preimplantation genetic diagnosis are possible [4, 19]. Views on the use of prenatal diagnosis differ between professionals and between families, and the decision is a personal one [4]. The AAO recommends genetic counselling after the results, by an experienced physician or a genetic counsellor, and its repetition in adulthood for diseases with childhood onset [15].

8.4 Contiguous Xq21 deletion and chromosomal rearrangements

In the contiguous Xq21 deletion, mothers and other women in the family may carry the deletion, sometimes with mild high-frequency hearing loss and signs of choroideremia carrier status; detection is by copy-number analysis (for example, MLPA), which has allowed asymptomatic carrier mothers to be identified; prenatal diagnosis of a cytogenetically visible deletion was described in 1987 [1, 3]. An affected woman with an X-autosome translocation disrupting CHM has reproductive risks specific to the rearrangement, which requires a karyotype and specific counselling [1, 27].

9. Special Considerations: Population Diversity and Epidemiology

9.1 What the estimates measure

Prevalence estimates for choroideremia count mainly affected men and do not include heterozygotes with signs, and they differ in method [7]. GeneReviews estimates a prevalence between 1:50,000 and 1:100,000, based on Khan et al. [4]. Orphanet records a European point prevalence in the 1–9/100,000 class, with a mean value of 2/100,000, taken from a 2014 EMA assessment [2]; the value of 2/100,000 corresponds to the upper end of the GeneReviews estimate [7]. The STAR authors quote 0.5–2/100,000 and consider the disease probably underdiagnosed because of its early resemblance to RP [9].

9.2 Founder populations

Northern Finland holds the largest founder population described. In 1980 about 58 Finnish cases had been identified, almost all in the north; Kärnä traced 111 patients and 188 carriers in four northern families and one from Savo, and the Salla family, spanning 13 generations, represented about one fifth of the patients known worldwide [1, 5]. The Salla variant (NM_000390.4:c.1609+2dup) is specific to that family [5, 31]. The study by Lähteenoja et al. shows that the current prevalence in northern Finland is much higher than the global estimate, and that CHM is among the most frequent causal genes of retinal dystrophies in the region, along with FZD4, RPGR, RS1, CERKL and CLRN1 [8]. The GeneReviews data on deletions and duplications are based mainly on people of northern European ancestry and on Chinese cohorts [4]. A German study cited by Lähteenoja et al. achieved high molecular diagnostic rates for choroideremia with targeted next-generation sequencing [8]. Danish (6 variants in 12 families), Swedish, Italian (a variant in 19 of 20 probands) and Mexican series have been described [1, 5].

9.3 Portugal and Brazil

The sources consulted do not include Portuguese epidemiological data on choroideremia. In Portugal, groups at the Champalimaud Foundation and the NOVA Medical School, in Lisbon, carry out preclinical research on REP-1 biology in the RPE and on delivery of CHM mRNA to the retina [68, 69]. In Brazil, one Brazilian patient had a synonymous CHM variant with an effect on splicing [30], and two STAR participants were followed locally in Brazil during the pandemic [9]. The absence of data in the sources consulted does not demonstrate the absence of national studies.

9.4 Special populations

Heterozygous women form the largest special population, with phenotypes ranging from asymptomatic mottling to male-pattern disease (section 3.6) [39–41]. Patients with a contiguous Xq21 deletion combine a dual sensory deficit, visual and auditory, with intellectual disability, which requires coordination between ophthalmology, otorhinolaryngology, audiology and neurodevelopment [3, 4].

9.5 Reference networks and patient associations

10. Illustrative Clinical Case

Palejwala, Lauer and Weleber, of the Casey Eye Institute (Portland, Oregon, USA), described in 2014 a 13-year-old Caucasian boy with choroideremia diagnosed at the age of 7, followed regularly in the ophthalmic genetics service, who had maintained acuity of 20/20 in both eyes until he presented with subacute loss of central acuity in the left eye [12]. On examination, acuity was 20/20 in the right eye and 20/100 in the left, with no afferent pupillary defect and normal tonometry and anterior segment. The fundus showed diffuse peripheral chorioretinal atrophy in both eyes, with the fovea spared on the right; in the left macula there was an elevated, greyish-white subfoveal membrane, with a small adjacent punctate subretinal haemorrhage (figure 1, ultra-wide-field fundus photograph: https://pmc.ncbi.nlm.nih.gov/articles/PMC4159397/figure/f1-opth-8-1675/) [12]. SD-OCT (figure 2: https://pmc.ncbi.nlm.nih.gov/articles/PMC4159397/figure/f2-opth-8-1675/) and fluorescein angiography (figure 3: https://pmc.ncbi.nlm.nih.gov/articles/PMC4159397/figure/f3-opth-8-1675/) showed a subretinal membrane with late staining and slight leakage, consistent with a neovascular membrane of mild activity, and, outside the central macula, diffuse absence of the choriocapillaris and RPE in both eyes [12]. The report does not state the CHM variant.

Given the loss of acuity and the neovascular lesion, anti-VEGF treatment was proposed: the patient received 10 intravitreal injections of bevacizumab 1.25 mg (Avastin®) over 12 months, an off-label use [12]. At the last examination, acuity was 20/150, the clinical examination was stable and SD-OCT showed slight anatomical improvement; the infrared images before and after treatment are in figure 4 (https://pmc.ncbi.nlm.nih.gov/articles/PMC4159397/figure/f4-opth-8-1675/) [12]. The authors attributed the limited response to the subretinal fibrosis already present and interpreted the change from 20/100 to 20/150 as fluctuation in acuity [12].

The case illustrates three points. First, early central loss in choroideremia is not always atrophy: in the second and third decades, the still-preserved submacular choriocapillaris allows new vessels to form through breaks in the degenerated Bruch membrane, by a mechanism compared to that of pathological myopia [12]. Second, the membranes are small, poorly exudative and may regress spontaneously, and visual loss may appear only with fibrosis; the authors suggest that anti-VEGF may reduce the activity of the membrane, but that, once scarring has occurred, the visual prognosis is limited [12]. Third, the same complication has been described in heterozygotes aged 27 and 46 years with «salt and pepper» RPE atrophy in the mid-periphery [12]. The injection schedule of the report is not a routine recommendation: there is no trial in choroideremia, the dose coincides with that of the European SmPC of bevacizumab for neovascular age-related macular degeneration in adults, and the decision to treat requires confirmation of lesion activity by imaging [12, 50]. In an adult, CNV of choroideremia falls within the European indication of ranibizumab for CNV of any cause; at the age of the patient described, any intravitreal anti-VEGF remains off-label (section 7.4) [14].

11. Prognosis and Long-Term Follow-Up

11.1 Prognosis

In men, central acuity remains good until about the age of 39 years and then declines by about 1.2 letters per year, with legal blindness (≤35 letters) at a median age of 66 years in the worse eye and 72 years in the better eye [10]. The area of preserved autofluorescence decreases by about 10% per year, and kinetic and static perimetry lose 12–20% per year, while acuity scarcely changes over 4 years in patients with a preserved paracentral ellipsoid zone [26]. Variability between individuals is wide, even within a family [1], and there are no established genotype-phenotype correlations [4]. Vision-related quality of life declines measurably over 2 to 4 years [26]. Heterozygotes with a granular phenotype remain stable; those with atrophic phenotypes progress [40, 41]. In the contiguous Xq21 deletion, the overall prognosis depends on the associated intellectual and hearing impairment [3]. The gene therapy studied up to 2026 has not demonstrably altered the prognosis in a phase 3 trial [9, 25].

11.2 Surveillance table

GeneReviews recommends regular ophthalmological examination, kinetic visual fields and SD-OCT especially when central vision is affected and cystoid macular oedema is suspected, without setting an interval [4]; the AAO sets an interval of 1 to 2 years for class 3 [15]. Children may need adapted regimens or sedation [15]. Patients should seek urgent examination for sudden loss of central vision or symptoms of retinal detachment, complications described in choroideremia [12, 35].

12. Summary and Final Considerations

Choroideremia is an X-linked chorioretinal dystrophy caused by loss of function of REP-1, the protein that presents Rab GTPases to Rab geranylgeranyl transferase; redundancy with other component A proteins is consistent with a disease limited to the eye, where degeneration of the RPE, photoreceptors and choriocapillaris progresses from the mid-periphery towards the centre. Men have night blindness in the first decade, field loss in the second and retain acuity until about 39 years, with legal blindness at a median age between 66 and 72 years; heterozygotes almost always have signs on autofluorescence, and a fraction develop progressive atrophic disease, through skewed X inactivation or mosaicism. Diagnosis requires sequencing of CHM with copy-number analysis, and, when this is negative in the face of a typical phenotype, RNA analysis and REP-1 immunoblot identify intronic variants, retrotransposons and rearrangements that sequencing does not detect. The main differential diagnosis is with RP, X-linked RP, gyrate atrophy (raised ornithine) and Usher syndrome, and association with deafness, intellectual disability or malformations points to a contiguous Xq21 deletion. Follow-up is every 1 to 2 years with multimodal imaging and perimetry, according to the AAO, with attention to cataract, cystoid macular oedema, retinal detachment, which is more frequent than in RP, and the rare choroidal neovascularisation of young patients. Subretinal gene therapy with AAV2-REP1 showed signs of stabilisation in the early trials and a modest effect in the 2026 meta-analysis, but the phase 3 STAR trial did not meet its primary endpoint, commercial development was halted, and in October 2026 there is no approved treatment in the USA or the European Union; natural history studies point to the area of preserved autofluorescence as an endpoint for future trials in less advanced disease, and mRNA delivery, optogenetics and molecular photoswitches are at preclinical or early stages. Genetic counselling, testing of the maternal line and low-vision rehabilitation are permanent components of management.

13. References

1. Online Mendelian Inheritance in Man. #303100 Choroideremia; CHM. Baltimore: Johns Hopkins University. Available from: https://omim.org/entry/303100
2. Orphanet. Choroideremia (ORPHA:180). Available from: https://www.orpha.net/en/disease/detail/180
3. Online Mendelian Inheritance in Man. #303110 Choroideremia, deafness, and mental retardation. Available from: https://omim.org/entry/303110
4. MacDonald IM, Hume S, Zhai Y, Xu M. Choroideremia. 2003 Feb 21 [updated 2021 Mar 4]. 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/NBK1337/
5. Online Mendelian Inheritance in Man. *300390 CHM Rab escort protein; CHM. Available from: https://omim.org/entry/300390
6. VariantValidator. Normalisation of CHM variants on NM_000390.4 (MANE Select), GRCh38. Available from: https://variantvalidator.org
7. GeneGuidance. CHM — Rab escort protein / Choroideremia (full record, consulted on 2026-09-26).
8. Lähteenoja L, Ohtonen P, Falck A, Rahikkala EJ. Characterisation and prevalence of inherited retinal diseases in the Finnish population reveals enrichment of population-specific phenotypes and causative variants. Br J Ophthalmol. 2025;109(8):852-7. doi:10.1136/bjo-2025-327427
9. MacLaren RE, Fischer MD, Gow JA, Lam BL, Sankila EK, Girach A, et al. Subretinal timrepigene emparvovec in adult men with choroideremia: a randomized phase 3 trial. Nat Med. 2023;29(10):2464-72. doi:10.1038/s41591-023-02520-3
10. Shen LL, Ahluwalia A, Sun M, Young BK, Grossetta Nardini HK, Del Priore LV. Long-term natural history of visual acuity in eyes with choroideremia: a systematic review and meta-analysis of data from 1004 individual eyes. Br J Ophthalmol. 2021;105(2):271-8. doi:10.1136/bjophthalmol-2020-316028
11. Khan KN, Islam F, Moore AT, Michaelides M. Clinical and genetic features of choroideremia in childhood. Ophthalmology. 2016;123(10):2158-65. doi:10.1016/j.ophtha.2016.06.051
12. Palejwala NV, Lauer AK, Weleber RG. Choroideremia associated with choroidal neovascularization treated with intravitreal bevacizumab. Clin Ophthalmol. 2014;8:1675-9. doi:10.2147/OPTH.S68243
13. European Medicines Agency. Orphan designations EU/3/14/1278, EU/3/14/1290 and EU/3/15/1482 (choroideraemia). Available from: https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu-3-14-1278; https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu3141290; https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu3151482
14. European Medicines Agency. Lucentis (ranibizumab): EPAR – summary of product characteristics. First authorisation on 22-01-2007. Available from: https://www.ema.europa.eu/en/documents/product-information/lucentis-epar-product-information_en.pdf
15. Duncan JL, Branham K, Birch DG, et al.; American Academy of Ophthalmology. Guidelines on clinical assessment of patients with inherited retinal degenerations – 2022. San Francisco: AAO; 2022 Oct. Available from: https://www.aao.org/education/clinical-statement/guidelines-on-clinical-assessment-of-patients-with
16. European Reference Network for Rare Eye Diseases (ERN-EYE). Available from: https://www.ern-eye.eu
17. Choroideremia Research Foundation. Available from: https://curechm.org
18. Retina Brasil. Available from: https://retinabrasil.org.br
19. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 302–5 (Retinal receptor dystrophies).
20. Cremers FP, van de Pol DJ, van Kerkhoff LP, Wieringa B, Ropers HH. Cloning of a gene that is rearranged in patients with choroideraemia. Nature. 1990;347(6294):674-7. doi:10.1038/347674a0
21. Seabra MC, Brown MS, Goldstein JL. Retinal degeneration in choroideremia: deficiency of rab geranylgeranyl transferase. Science. 1993;259(5093):377-81. doi:10.1126/science.8380507
22. García-Cazorla Á, Dionisi-Vici C, Saudubray JM. Disorders of cellular trafficking. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 44, Table 44.2.
23. Abdalla Elsayed MEA, Cehajic-Kapetanovic J, MacLaren RE. Gene therapy for choroideremia: progress, potential and pitfalls. Expert Opin Biol Ther. 2025;25(3):257-63. doi:10.1080/14712598.2025.2459850
24. MacLaren RE, Groppe M, Barnard AR, Cottriall CL, Tolmachova T, Seymour L, et al. Retinal gene therapy in patients with choroideremia: initial findings from a phase 1/2 clinical trial. Lancet. 2014;383(9923):1129-37. doi:10.1016/S0140-6736(13)62117-0
25. Chen KY, Chan HC, Chan CM. Efficacy and safety of AAV-mediated gene therapy for choroideremia: a systematic review and meta-analysis. EClinicalMedicine. 2026;95:103923. doi:10.1016/j.eclinm.2026.103923
26. Birch DG, Bernstein PS, MacDonald IM, Stout T, Liao DS, Locke KG, et al. A 48-month, prospective, multicenter, natural history study of choroideremia. Ophthalmol Sci. 2026;6(10):101361. doi:10.1016/j.xops.2026.101361
27. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 50–1 (X-linked recessive inheritance).
28. Hayashi T, Maki H, Higa N, Mizobuchi K, Maruyama T, Hayashi T, et al. Somatic mosaicism and skewed X-chromosome inactivation associated with male-pattern choroideremia phenotype in a female patient. Doc Ophthalmol. 2026 Sep 16. doi:10.1007/s10633-026-10149-z
29. Moosajee M, Ramsden SC, Black GC, Seabra MC, Webster AR. Clinical utility gene card for: choroideremia. Eur J Hum Genet. 2014;22(4):572. doi:10.1038/ejhg.2013.183
30. da Palma MM, Motta FL, Gomes CP, Salles MV, Pesquero JB, Sallum JMF. Synonymous variant in the CHM gene causes aberrant splicing in choroideremia. Invest Ophthalmol Vis Sci. 2020;61(2):38. doi:10.1167/iovs.61.2.38
31. National Center for Biotechnology Information. ClinVar: VCV000056224 (NM_000390.4:c.1609+2dup) and VCV000039809 (NM_000390.4:c.1520A>G). Available from: https://www.ncbi.nlm.nih.gov/clinvar/
32. Fry LE, Patrício MI, Williams J, Aylward JW, Hewitt H, Clouston P, et al. Association of messenger RNA level with phenotype in patients with choroideremia: potential implications for gene therapy dose. JAMA Ophthalmol. 2020;138(2):128-35. doi:10.1001/jamaophthalmol.2019.5071
33. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 122–5 (Deafness in early childhood).
34. MacLaren RE, Lam BL, Fischer MD, Holz FG, Pennesi ME, Birch DG, et al. A prospective, observational, non-interventional clinical study of participants with choroideremia: the NIGHT study. Am J Ophthalmol. 2024;263:35-49. doi:10.1016/j.ajo.2024.01.022
35. Al-Moujahed A, Haque ME, Mahajan VB, Huckfeldt R, Pershing S. Epidemiology of inherited retinal diseases in the United States: IRIS® Registry (Intelligent Research in Sight) analysis. Ophthalmic Surg Lasers Imaging Retina. 2026;57(4):257-62. doi:10.3928/23258160-20260310-03
36. Dossett JP, Clavell CI, Eid A, Pumariega N, Ghorayeb G. Anatomic and functional recovery following surgical repair of full-thickness macular hole in choroideremia. J Vitreoretin Dis. 2026;10(5):1055-9. doi:10.1177/24741264261445182
37. Bass BA, Kennedy BJ, Flaxel CJ, Kroeger ZA. Retinoschisis-associated retinal detachment in choroideremia: a surgical conundrum. Retin Cases Brief Rep. 2026 Oct 14. doi:10.1097/ICB.0000000000002001
38. Abdalla Elsayed MEA, Barone V, MacLaren RE. Vitrectomy and internal limiting membrane peeling for extensive macular schisis in choroideremia. Graefes Arch Clin Exp Ophthalmol. 2026;264(9):2757-62. doi:10.1007/s00417-026-07162-y
39. Jauregui R, Park KS, Tanaka AJ, Cho A, Paavo M, Zernant J, et al. Spectrum of disease severity and phenotype in choroideremia carriers. Am J Ophthalmol. 2019;207:77-86. doi:10.1016/j.ajo.2019.06.002
40. Gocuk SA, Ayton LN, Edwards TL, McGuinness MB, MacLaren RE, Taylor LJ, et al. Longitudinal assessment of female carriers of choroideremia using multimodal retinal imaging. Br J Ophthalmol. 2025;109(2):293-9. doi:10.1136/bjo-2024-325578
41. Han X, Yu Y, Ding J, Sun Z, Li H, Zou X, et al. Longitudinal and cross-sectional study of retinal phenotypes and visual function in choroideremia carriers: a new grading system. Eye Vis (Lond). 2026;13(1). doi:10.1186/s40662-026-00476-2
42. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 404–8 (Kearns-Sayre syndrome).
43. Baumgartner MR, Valle D, Dionisi-Vici C. Disorders of ornithine and proline metabolism. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 21, section 21.1. doi:10.1007/978-3-662-63123-2_21
44. Stone EM, Aldave AJ, Drack AV, MacCumber MW, Sheffield VC, Traboulsi E, et al. Recommendations for genetic testing of inherited eye diseases: report of the American Academy of Ophthalmology task force on genetic testing. Ophthalmology. 2012;119(11):2408-10. doi:10.1016/j.ophtha.2012.05.047
45. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 518–9 (Retinitis pigmentosa).
46. U.S. Food and Drug Administration, Office of Orphan Products Development. Search Orphan Drug Designations and Approvals: designations for «choroideremia» (12/09/2013, 05/11/2014, 12/10/2017). Accessed on 31 Oct 2026. Available from: https://www.accessdata.fda.gov/scripts/opdlisting/oopd/
47. ClinicalTrials.gov. Registrations NCT03496012, NCT03507686, NCT03584165, NCT03359551, NCT02994368, NCT04483440, NCT02407678, NCT02341807, NCT01461213, NCT02671539, NCT02077361, NCT02553135, NCT06460844, NCT05282953, NCT06375239, NCT01654562, NCT05045703, NCT02435940. Available from: https://clinicaltrials.gov
48. Genead MA, McAnany JJ, Fishman GA. Topical dorzolamide for treatment of cystoid macular edema in patients with choroideremia. Retina. 2012;32(4):826-33. doi:10.1097/IAE.0b013e3182215ae9
49. U.S. Food and Drug Administration. Dorzolamide hydrochloride ophthalmic solution 2%: prescribing information (ANDA215660; label of 09/01/2026, reference Trusopt®). openFDA/DailyMed. Available from: https://dailymed.nlm.nih.gov
50. European Medicines Agency. Lytenava (bevacizumab gamma): EPAR – summary of product characteristics. Authorisation of 27-05-2024. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/lytenava
51. INFARMED. Infomed — Lytenava (bevacizumab) 25 mg/ml solution for injection, medicine details. Lisbon: INFARMED; accessed on 31 Oct 2026. Available from: https://extranet.infarmed.pt/INFOMED-fo/
52. U.S. Food and Drug Administration. Lucentis (ranibizumab injection) for intravitreal use: prescribing information (BLA125156; label of 21/11/2025). Available from: https://www.accessdata.fda.gov/scripts/cder/daf/
53. Novartis Biociências S.A. Lucentis® (ranibizumabe) solução injetável: bula para o profissional de saúde (VPS11), approved by ANVISA on 29/08/2025. Available from: https://portal.novartis.com.br/medicamentos/
54. INFARMED. Infomed — search «ranibizumab» (Lucentis, Ranivisio, Byooviz, Ranibizumab Midas, Ranluspec, Rexatilux). Lisbon: INFARMED; accessed on 31 Oct 2026. Available from: https://extranet.infarmed.pt/INFOMED-fo/
55. Agência Nacional de Vigilância Sanitária (ANVISA). Open data — medicines registered in Brazil (DADOS_ABERTOS_MEDICAMENTOS.csv, updated on 13/10/2026), active substances bevacizumabe and ranibizumabe. Accessed on 31 Oct 2026. Available from: https://dados.anvisa.gov.br/dados/DADOS_ABERTOS_MEDICAMENTOS.csv
56. Aleman TS, Huckfeldt RM, Serrano LW, Pearson DJ, Vergilio GK, McCague S, et al. Adeno-associated virus serotype 2-hCHM subretinal delivery to the macula in choroideremia: two-year interim results of an ongoing phase I/II gene therapy trial. Ophthalmology. 2022;129(10):1177-91. doi:10.1016/j.ophtha.2022.06.006
57. Edwards TL, Jolly JK, Groppe M, Barnard AR, Cottriall CL, Tolmachova T, et al. Visual acuity after retinal gene therapy for choroideremia. N Engl J Med. 2016;374(20):1996-8. doi:10.1056/NEJMc1509501
58. Xue K, Jolly JK, Barnard AR, Rudenko A, Salvetti AP, Patrício MI, et al. Beneficial effects on vision in patients undergoing retinal gene therapy for choroideremia. Nat Med. 2018;24(10):1507-12. doi:10.1038/s41591-018-0185-5
59. Dimopoulos IS, Hoang SC, Radziwon A, Binczyk NM, Seabra MC, MacLaren RE, et al. Two-year results after AAV2-mediated gene therapy for choroideremia: the Alberta experience. Am J Ophthalmol. 2018;193:130-42. doi:10.1016/j.ajo.2018.06.011
60. Lam BL, Davis JL, Gregori NZ, MacLaren RE, Girach A, Verriotto JD, et al. Choroideremia gene therapy phase 2 clinical trial: 24-month results. Am J Ophthalmol. 2019;197:65-73. doi:10.1016/j.ajo.2018.09.012
61. Fischer MD, Ochakovski GA, Beier B, Seitz IP, Vaheb Y, Kortuem C, et al. Efficacy and safety of retinal gene therapy using adeno-associated virus vector for patients with choroideremia: a randomized clinical trial. JAMA Ophthalmol. 2019;137(11):1247-54. doi:10.1001/jamaophthalmol.2019.3278
62. Cehajic-Kapetanovic J, Bellini MP, Taylor LJ, Yusuf IH, Soomro T, da Cruz L, et al. Gene therapy for choroideremia using an adeno-associated viral vector encoding Rab escort protein 1: the REGENERATE open-label trial. Efficacy Mech Eval. 2024 May. p. 1-59. doi:10.3310/WARA5730
63. Biogen Inc. Annual report (Form 10-K) for the fiscal year ended December 31, 2021. Filed 2022-02-03. Available from: https://www.sec.gov/Archives/edgar/data/875045/000087504522000007/biib-20211231.htm
64. MacLaren RE, Audo I, Fischer MD, Huckfeldt RM, Lam BL, Pennesi ME, et al. An open-label phase II study assessing the safety of bilateral, sequential administration of retinal gene therapy in participants with choroideremia: the GEMINI study. Hum Gene Ther. 2024;35(15-16):564-75. doi:10.1089/hum.2024.017
65. 4D Molecular Therapeutics Inc. Annual report (Form 10-K) for the fiscal year ended December 31, 2022. Filed 2023-03-15. Available from: https://www.sec.gov/Archives/edgar/data/1650648/000095017023008142/fdmt-20221231.htm
66. 4D Molecular Therapeutics Inc. Annual report (Form 10-K) for the fiscal year ended December 31, 2024. Filed 2025-02-28. Available from: https://www.sec.gov/Archives/edgar/data/1650648/000095017025030056/fdmt-20241231.htm
67. Casson RJ, Daniels E, Barras CD, Dwyer A, Strem BM, Wykoff CC, et al. Intravitreal photoswitch therapy in advanced retinitis pigmentosa: a phase 1 open-label trial. Nat Med. 2026;32(5):1865-70. doi:10.1038/s41591-026-04317-6
68. de Lemos L, Antas P, Carvalho C, Castro M, Bruno D, Fonseca AF, et al. mRNA delivery to the retina restores REP1 function in choroideremia. Mol Ther Adv. 2026;34(3):201768. doi:10.1016/j.omta.2026.201768
69. Coelho R, Antas P, Fonseca AF, Oliveira D, Carvalho C, Pedro ML, et al. Loss of REP-1 in retinal pigment epithelial cells leads to impaired phagosome processing and altered lysosomal pathway function. Mol Biol Cell. 2025;36(9):ar116. doi:10.1091/mbc.E24-11-0497
70. Abbass NJ, Yazji I, Allan KC, Kaelber DC, Talcott KE, Singh RP. Trends and disparities in the incidence and prevalence of inherited retinal diseases in the United States. Am J Ophthalmol. 2025;279:165-73. doi:10.1016/j.ajo.2025.07.021
71. Foundation Fighting Blindness. Available from: https://www.fightingblindness.org
72. Retina International. Available from: https://retina-international.org