Congenital Aniridia and the PAX6 Spectrum

Literature and regulatory information updated: 6 October 2026.

PAX6 aniridia syndrome (classical aniridia and PAX6-related ocular and systemic phenotypes, including deletions of the 3′ regulatory region at 11p13 involving ELP4) and WAGR spectrum disorder (contiguous 11p13 deletion involving WT1 and PAX6)

0. Initial Reference Table

1. Introduction

Congenital aniridia is a dominantly inherited panocular malformation named after its most visible sign, partial or complete absence of the iris, but its functional burden arises mainly from foveal hypoplasia, present from birth, and progressive complications of the cornea, lens and iridocorneal angle that accumulate throughout life [1, 9]. The current GeneReviews designation, “PAX6 aniridia syndrome”, replaced “PAX6-related aniridia” to emphasise that the condition includes extraocular manifestations, such as abnormalities of central auditory processing, smell, sleep, neurodevelopment and glucose metabolism [1]. The distinction carrying the greatest weight in genetic consultation is the boundary between aniridia caused by an alteration confined to PAX6 and a contiguous 11p13 deletion also involving WT1, WAGR spectrum disorder, because only the latter confers a risk of Wilms tumour and chronic kidney disease [1, 3].

Classical aniridia was first described in 1818 by Barratta [2]. In 1964, Miller, Fraumeni and Manning, reviewing the records of 440 children with Wilms tumour, found six (1.4%) with aniridia, an association exceeding chance given that the incidence of aniridia was then estimated at 1:50,000 live births; all had intellectual disability or global delay, three had microcephaly and one had contralateral renal hypoplasia [2, 16]. In 1978, Riccardi and colleagues identified interstitial deletions of the short arm of chromosome 11 in three unrelated patients with aniridia, ambiguous genitalia and intellectual disability, one of whom had Wilms tumour, and proposed that the tumour was one possible consequence of the same chromosomal alteration [17]. The bibliography of the Smith chapter also records the work of Francke and colleagues (1979), providing evidence of a specific 11p13 deletion, and Yunis and Ramsay (1980), concerning familial occurrence of the association with an 11p13-14.1 deletion [18].

Positional cloning of WT1 in 1990 and PAX6 in 1991 showed that the two genes lie less than one megabase apart, establishing WAGR syndrome as the paradigm of contiguous gene syndromes [2]. Ton and colleagues cloned, in the region of the aniridia locus, a complementary DNA wholly or partly deleted in two patients, expressed in affected tissues and encoding a polypeptide with a paired domain, a homeodomain and a serine- and threonine-rich carboxy-terminal domain [19]. Simultaneous analysis of the murine Small eye (Sey) alleles, described in 1967 as microphthalmia in heterozygotes and subsequently with iris hypoplasia, cataract and corneal opacification, confirmed the causal relationship and revealed the gene's dosage sensitivity [2]. In 2021, analysis of 91 participants in the international patient registry led Duffy and colleagues to propose the term “WAGR spectrum disorder”, adopted by GeneReviews in 2026 [3, 8].

Clinical knowledge has recently been organised in three 2026 documents: the full revision of the GeneReviews entry for PAX6 aniridia syndrome (June 2026), which states that no published clinical practice guidelines exist and relies on the authors' experience; the European guidelines of the COST action ANIRIDIA-NET, prepared with the Aniridia-Europa/Geniris working group from a literature review between 2009 and May 2025 and from the French and Swedish national protocols; and the Japanese guideline, developed according to the Minds 2017 manual, with clinical questions and formal grading of the strength of recommendations [1, 9, 10]. For WAGR, the GeneReviews entry published in March 2026 adopts the screening programme proposed by the patient registry, and European cancer surveillance follows the 2021 recommendations of the SIOP-Europe Host Genome Working Group and the SIOP Renal Tumour Study Group [3, 8, 11]. In France, the Haute Autorité de Santé (HAS) published in October 2021 the PNDS (French National Diagnosis and Care Protocol) for WAGR syndrome, written by the OPHTARA rare diseases reference centre in ophthalmology with nephrologists, oncologists, paediatric neurologists and cytogeneticists, which refers ophthalmological detail to the 2019 PNDS for congenital aniridia [7, 20].

2. Molecular Genetics and Pathophysiology

2.1. The PAX6 gene, isoforms and nomenclature

PAX6 is located at 11p13 (hg38 chr11:31,784,792–31,811,353) and belongs to the PAX family of paired-box genes, which encode highly conserved transcription factors controlling organogenesis [1, 2]. The canonical protein has 422 amino acids (UniProt P26367-1) and contains two DNA-binding domains, the bipartite paired domain and the homeodomain, together with a proline-, serine- and threonine-rich carboxy-terminal transactivation domain; the alternative PAX6(5a) isoform, with 436 amino acids, includes 14 additional residues between Gln47 and Val48, encoded by exon 5a, which modulate DNA binding through the carboxy-terminal subdomain [2]. The coding exons are exons 4 to 13; the first three exons are non-coding [1, 2]. PAX6 is abundantly expressed in the developing eye and maintains postnatal expression in the cornea, conjunctiva, iris, ciliary body, lens and retina, as well as in the central nervous system and endocrine pancreas [1, 2].

This architecture has practical consequences for interpreting reports. The MANE Select transcript of PAX6 is NM_001368894.2, encoding the isoform containing exon 5a (NP_001355823.1), whereas much of the literature uses the numbering of the canonical 422-amino-acid isoform (NM_000280) [2, 21]. Thus, the variant described as NM_000280.3:c.178T>A p.(Tyr60Asn) corresponds in the MANE transcript to NM_001368894.2:c.220T>A p.(Tyr74Asn) and in the genome to NC_000011.10:g.31801740A>T (GRCh38) [21, 22]. For residues after Gln47, the difference in protein numbering between the two references is 14 amino acids; domain positions cited in this article, such as the paired domain at amino acids 4 to 130, follow the canonical numbering used by the sources [1, 2]. The report must specify the transcript and version, and conversion between numbering systems must be confirmed using a normalisation tool rather than calculated manually [21].

2.2. Mechanism: haploinsufficiency and dosage sensitivity

In the vast majority of cases, aniridia results from haploinsufficiency: heterozygous loss-of-function variants reduce the amount of functional protein below the threshold required for ocular development [1, 23]. PAX6 has a loss-of-function intolerance index (pLI) of 1.00 in ExAC; duplication of the entire gene causes ocular malformations other than aniridia, and biallelic loss of function is lethal in the perinatal period in humans and mice, demonstrating that both excessive and insufficient dosage are pathogenic [2]. The rare reported cases of biallelic loss-of-function variants presented with severe craniofacial anomalies, anophthalmia, an absent or malformed nose, absent adrenal glands, central nervous system malformations and, frequently, fetal or neonatal death; however, viable compound heterozygotes with aniridia or coloboma have occurred when one allele was a presumably hypomorphic missense variant [1, 2].

Tissue pathophysiology explains the natural history. Foveal hypoplasia reflects early arrest of foveal specialisation; in approximately 70% of patients it reaches grades 3 or 4 of the Thomas classification, consistent with interruption during centripetal cone migration, and the thickness of the outer layers of the foveal retina is the best predictor of visual acuity [24]. AAK results from functional limbal stem cell deficiency and abnormal healing: conjunctival epithelium and vessels progressively invade the cornea from the limbus, with disruption of the palisades of Vogt, opacification and neovascularisation [1, 9]. Recent studies of large European cohorts, involving more than 550 eyes, described a reduced sub-basal nerve plexus, keratocyte and endothelial abnormalities, infiltration by Langerhans cells, microRNA dysregulation (such as miR-204-5p and miR-138-5p) and an increased inflammatory response in limbal fibroblasts, pointing to molecular targets for future therapies [25]. Glaucoma arises mainly from trabecular meshwork dysgenesis and progressive obstruction of the angle by the iris stump, which migrates forwards and forms synechiae; absence of Schlemm's canal is another described mechanism [9]. In the pancreas, PAX6 is required for beta-cell identity and proinsulin processing through regulation of prohormone convertase 1/3, underpinning the association with glucose intolerance [24].

2.3. Variant spectrum and genotype–phenotype correlations

The PAX6 LOVD database recorded 472 unique variants, with no domain spared [2]. The most frequent are substitutions that create a premature termination codon (PTC), and the three most often reported, p.Arg240*, p.Arg317* and p.Arg203* (canonical numbering; 129 reports in LOVD), lie at CpG dinucleotides [2]. According to GeneReviews, 70–85% of variants associated with haploinsufficiency introduce a PTC (nonsense, frameshift and intronic or splicing variants) and are associated with classical aniridia with a high incidence of vision-reducing complications [1]. Missense variants, particularly in the paired domain (amino acids 4–130), are associated with atypical, milder or variable phenotypes, with more residual iris tissue and better vision, but also with non-aniridia phenotypes, such as microphthalmia, anophthalmia and coloboma, isolated foveal hypoplasia and congenital nystagmus [1]. Variants that extend the protein in the C-terminal domain (CTE) are associated with moderately severe aniridia, with poor vision, keratopathy and cataract, less glaucoma and more iris preservation, and rarely with exudative retinopathy and chorioretinal degeneration [1, 2].

The longitudinal cohort of Kit and colleagues (86 patients from 62 families, United Kingdom, 172 eyes, mean follow-up 16.3 years) shows the distribution in a clinical population: nonsense 16.3%, frameshift 24.4%, intronic/splicing 17.4%, CTE 16.3%, missense 22.1% and deletions 3.6% (two whole-gene deletions and one deletion of the 3′ regulatory region within ELP4); all missense variants were located in exons 5 and 6, which encode the paired domain [26]. The missense group had the mildest phenotype, and eyes that had never undergone surgery maintained better visual acuity [26]. Nevertheless, GeneReviews emphasises that the described correlations cannot predict severity in an individual [1]. In the cornea, PTC variants follow the classical age-dependent progression, multiexon or whole-gene deletions are associated with earlier and more rapid AAK, WAGR with the most severe forms (grades 2 to 4 already in childhood, in four patients in a German cohort), and deletions confined to the 3′ regulatory regions with absent or minimal keratopathy [24]. In a Russian cohort of 110 patients, 14 of 15 carriers of regulatory deletions had no keratopathy and generally had a mild phenotype without nystagmus or foveal hypoplasia, whereas two-thirds of missense variants produced phenotypes other than aniridia [24].

2.4. Long-range regulation, ELP4 and 11p13 deletions without WT1

PAX6 expression depends on conserved cis-regulatory elements, some hundreds of kilobases away [2, 5]. Translocation and inversion breakpoints in patients with aniridia situated 11, 22, 125 and 150 kb downstream of the gene (the latter within the SIMO element, an autoregulatory site) produced phenotypes identical to those caused by deletions or PTC variants, and a critical 244 kb transcriptional activation region was proposed [2]. 11p13 deletions encompassing genes downstream of PAX6, such as DCDC1, DPH4, IMMP1L and ELP4, can cause aniridia without affecting the coding sequence, by removing enhancers [1]. ELP4 encodes a subunit of the Elongator complex, but its clinical relevance is positional: its introns house PAX6 regulatory elements, and in November 2023 ClinGen classified its relationship with the entity “ocular dysgenesis caused by defects of PAX6 regulation” as moderate; OMIM lists “?aniridia 2” (617141) with a question mark, indicating a provisional molecular basis [5]. Microdeletions of the ELP4 region have been associated with language disorder, autism spectrum disorder (ASD) and intellectual disability [3, 5]. Contiguous 11p13 deletions including PAX6 and ELP4, but not WT1, can produce ocular and developmental phenotypes resembling WAGR, without a reported increase in Wilms tumour risk [3].

2.5. WAGR spectrum disorder: contiguous genes

All patients described with WAGR have an 11p13 deletion that includes WT1 and PAX6; haploinsufficiency of WT1, a transcription factor with a role in the differentiation of the kidney, urinary tract and gonadal progenitors and with tumour suppressor function, explains the risk of Wilms tumour, and that of PAX6 the aniridia [3]. The tumour follows the «two-hit» model: the constitutional deletion inactivates one allele and somatic loss of the second, in the kidney, initiates tumorigenesis [2]. Deletions range from 1.0 to 26.5 Mb in the series of Han and colleagues, and the size of the lost segment, especially distal to 11p13, contributes to variability in growth and cognition [6, 18]. The French PNDS gives the same range, with a mean of 11 Mb, and notes that deletions extending to 11p14–p12 may include, besides BDNF, LMO2, EXT2, ALX4, PRRG4 and TRIM44 [7]. Deletion of BDNF (11p14.1, approximately 4 Mb centromeric to WT1/PAX6) was present in 58% of 33 patients; at 10 years of age, 100% of carriers of a BDNF deletion were obese (95% CI, 77–100) versus 20% of the others (95% CI, 3–56), with serum BDNF concentrations approximately 50% lower and a critical region for childhood obesity less than 80 kb from exon 1 [6]. In the Duffy registry, BDNF deletion was reported by 27 of 54 valid respondents (50%) [8]. BDNF haploinsufficiency is also associated with lower adaptive behaviour, ASD, hyperphagia, reduced pain sensitivity and, possibly, sleep disturbance [3, 18].

Other genes in the region have been associated with the extended phenotype, with evidence limited by small samples: ELP4 (language disorder, ASD, epilepsy), PRRG4 (developmental delay and ASD), SLC1A2 (epilepsy) and ARL14EP (severe lens anomalies together with PAX6 deletion) [3]. Deletions extending to ALX4 and EXT2 cause concomitant Potocki-Shaffer syndrome, with enlarged parietal foramina and multiple osteochondromas; PHF21A, associated with ASD, epilepsy and craniofacial anomalies, is deleted more rarely in these combined cases [3]. GeneReviews does not identify clinically relevant genotype–phenotype correlations among the causal deletions, except for the effects attributed to the additional genes [3]. The French PNDS adds two hypotheses based on small series: that the risk of Wilms tumour depends on joint loss of WT1 and LMO2 (11p13), from a Russian study, and that PRRG4 contributes to autistic traits; both coexist with the second-hit model on the non-deleted WT1 allele and do not alter surveillance [7]. The same document recalls that WT1 acts on the SRY promoter, which explains the genital anomalies, and that only 5–10% of isolated Wilms tumours have a WT1 variant [7]. Mosaic 11p13 deletions are rare and, in the cases described, the level of mosaicism was low enough to escape conventional CMA [3].

2.6. Allelic disorders and related entities

Heterozygous PAX6 variants cause, besides aniridia, entities with their own OMIM entries: microphthalmia/coloboma 12 (120200), optic nerve coloboma and morning glory anomaly (120430), foveal hypoplasia 1 (136520), hereditary keratitis (148190), bilateral optic nerve hypoplasia (165550) and anterior segment dysgenesis 5, including Peters anomaly (604229) [1, 4]. Missense variants in the paired domain have been found in cases of Peters anomaly, and heterozygous variants in families with coloboma, morning glory, optic nerve hypoplasia or aplasia and persistent hyperplastic primary vitreous [27, 28]. Two missense variants of the PAX6(5a) isoform were described in Japanese families with congenital corneal opacification, with or without foveal hypoplasia and cataract [2]. Intragenic WT1 variants, in turn, cause WT1 disorder (Denys–Drash/Frasier spectrum), without aniridia, and are discussed in the differential diagnosis [3, 5].

3. Clinical Spectrum and Manifestations by System

3.1. Ocular manifestations of PAX6 aniridia syndrome

Classical aniridia affects the iris, cornea, iridocorneal angle, lens, fovea and optic nerve; severity varies between and within families, but the two eyes of the same individual tend to be similar [1]. Visual acuity usually lies between 20/100 and 20/200, and the European guidelines describe values of 20/100 to 20/1000 in adulthood; the initial reduction arises mainly from foveal hypoplasia, whereas cataract, glaucoma and corneal opacification cause progressive visual loss [1, 9]. Nevertheless, GeneReviews emphasises that most patients retain useful vision with appropriate ophthalmological follow-up [1].

Iris. Hypoplasia ranges from an almost normal iris, with loss of surface architecture or only transillumination, to apparent absence, with an iris stump visible only on gonioscopy, anterior segment optical coherence tomography (AS-OCT) or ultrasound biomicroscopy (UBM) [1]. Partial coloboma-like defects, eccentric or distorted pupils and ectropion uveae may occur [1, 9]. In some patients the iris is present and almost intact, making the term “aniridia” inappropriate and diagnosis dependent on foveal hypoplasia and genetic testing [24].

Fovea and nystagmus. Foveal hypoplasia is present in most patients, with prevalences of 79–92% in the European guidelines and over 90% in a 2023 review; it manifests as a reduced foveal reflex, macular hypopigmentation and vessels crossing the foveal avascular zone [9, 24]. Approximately 70% of patients with a PAX6 variant and a minimally affected iris have severe foveal hypoplasia, so iris severity does not predict foveal severity [24]. Nystagmus usually develops by six weeks of life, has features of infantile nystagmus syndrome, frequently with a vertical component, and was recorded in 87.2% of eyes in Kit's British cohort [1, 24, 26]. Unlike albinism, foveal hypoplasia in aniridia is not accompanied by abnormal chiasmal decussation [2].

Cornea and ocular surface. AAK results from limbal stem cell deficiency and ranges from mild peripheral vascularisation to vascularisation, opacification and keratinisation of the entire cornea; insufficient tear production exacerbates surface disease [1]. The European guidelines estimate dry eye in 56–96%, AAK (opacification and neovascularisation indicating limbal deficiency) in 78–96%, and minimal keratopathy, with neurotrophic and inflammatory changes, in 100% of patients; keratopathy asymmetry between eyes occurs in approximately 30% [9]. Central corneal thickness is increased, typically approximately 600 μm in early stages and up to 1 mm in advanced stages, requiring intraocular pressure (IOP) correction using pachymetry [9]. Signs of AAK usually become evident in early adulthood, and keratopathy is the main cause of progressive visual loss at older ages [2]. Microcornea (diameter below 10 mm) occurs rarely and megalocornea very rarely [1].

Lens. Congenital opacities, particularly polar opacities, are frequent, sometimes with persistence of the tunica vasculosa lentis or pupillary membrane remnants; they rarely require extraction in childhood, but visually significant opacities develop in 50–85% of patients, often during adolescence or early adulthood [1]. The European guidelines report cataract prevalences up to 90% and describe adult “wagon-wheel” cataract as characteristic of PAX6 aniridia; other anomalies include lens coloboma, microspherophakia, posterior lenticonus and subluxation, the latter uncommon [1, 9]. The anterior capsule is thinner (less than 8 μm) and fragile, and zonular fragility complicates surgery [9]. In the differential diagnosis of congenital cataract, association with other anterior segment malformations, such as aniridia, distinguishes these forms from isolated cataracts [29].

Intraocular pressure and glaucoma. Glaucoma usually begins in late childhood or adulthood; congenital glaucoma with buphthalmos is rare in PAX6 aniridia [1, 9]. Estimated cumulative prevalence reaches two-thirds (GeneReviews), 43–70% (European guidelines, with approximately 15% below age 10 and a mean age at diagnosis of approximately 15 years, probably underestimated) and 50–75% (Japanese guideline) [1, 9, 10]. Ocular hypertension without glaucomatous damage is also frequent and constitutes a risk factor [1]. Cataract surgery in childhood adds an independent risk of secondary glaucoma [2].

Optic nerve, retina and other findings. Optic nerve hypoplasia occurs in up to 10% according to GeneReviews and in 11–29% according to the European guidelines, which also describe a pale fundus in 86% [1, 9]. Optic nerve coloboma, retinal detachment (associated with high myopia or previous intraocular surgery) and, very rarely, exudative vascular retinopathy or chorioretinal degeneration may occur [1]. Refractive errors range from high myopia to high hypermetropia; amblyopia may be refractive or strabismic, and some patients have microphthalmia and iris, chorioretinal or disc coloboma [1]. Eyelid ptosis is part of the clinical picture described in reviews and the French national protocol [2, 20]. Aniridic fibrosis syndrome (AFS), which is rare, develops after multiple intraocular operations, with a retrolenticular and retrocorneal fibrotic membrane originating at the iris root, which may displace or entrap the intraocular lens and cause corneal decompensation [1].

The following table brings together reported frequencies and their respective denominators, which are not interchangeable: some are lifetime cumulative prevalences per patient, whereas others are proportions of eyes in a cohort with defined follow-up.

Sources: [1, 9, 10, 26]. In the series of 94 patients from a North American centre (median age at presentation 19 years), glaucoma affected 52% of eyes and keratopathy 65%, with two-thirds of patients having undergone at least one intraocular operation [30].

3.2. Extraocular manifestations of PAX6 aniridia syndrome

GeneReviews includes central auditory processing difficulties in the phenotype, arising from impaired interhemispheric transfer, which hinder speech comprehension despite normal peripheral hearing and have particular consequences for people with reduced vision [1]. Hyposmia or anosmia is related to hypoplasia or agenesis of the olfactory bulbs and tracts and should be sought during evaluation, as noted in the Oxford chapters on anterior segment malformations and glaucoma [1, 27, 31]. Sleep disturbance is attributed to absence or hypoplasia of the pineal gland, with reduced melatonin secretion, and lower light exposure [1, 3]. Brain magnetic resonance imaging (MRI) may show absence or hypoplasia of the anterior commissure, a reduced posterior commissure, chiasmal atrophy, grey matter abnormalities in the anterior cingulate, cerebellum and temporal and occipital lobes, white matter deficits and reduced corpus callosum volume, an absent pineal gland and hypoplastic olfactory bulbs [1]. In a familial series cited by the Japanese guideline, MRI showed pineal deficiency in 30%, severe cerebral dysplasia in 10%, a posterior commissure defect in 10% and chiasmal or callosal atrophy in 10% [10].

The impact of early visual impairment affects all developmental domains, although profound visual impairment is rare in PAX6 aniridia syndrome; many patients have average cognitive ability, and delay persisting beyond age five should be recognised as a learning difficulty or, if accompanied by adaptive deficits, as intellectual disability [1]. ASD, attention deficit hyperactivity disorder (ADHD), language difficulties, behavioural dysregulation, depression and anxiety are frequent, irrespective of cognitive ability [1].

Regarding metabolism, GeneReviews includes obesity and glucose intolerance or diabetes [1]. In Kit's cohort, 12.8% of patients had type 2 diabetes, approximately twice the estimated prevalence in the United Kingdom, and 23.3% had obesity, below the British population prevalence [24, 26]. An earlier North American survey estimated diabetes at 7%, a figure Hall and colleagues compare with 4.5% in the population, cautioning that the populations are not directly comparable [2]. In the Homburg Registry (Germany; 337 patients, mean age 22 years), 46% had at least one systemic disease: obesity 8.6%, thyroid disease 8.3%, hypertension 7.7%, intellectual disability 6.5%, diabetes 5.6%, auditory perception disorder or language delay 4.7% and epilepsy 3.6%; among the 187 with genetic testing, 174 (93%) had a PAX6 variant, including 20 with WAGR(O) [32]. A synthesis of series cited by the Japanese guideline also reports dental dysplasia (35%), musculoskeletal anomalies (13%), asthma (12%), depression (12%), infertility (11%), biliary disease (8%), hypertension (8%), diabetes (7%), hyposmia (5%) and pancreatitis (1%) [10]. These proportions come from ophthalmological referral centres and are not population estimates.

3.3. WAGR spectrum disorder

Aniridia and the eye. Aniridia is almost always present: 81 of 83 registry participants (97.6%), and the two without aniridia had amblyopia [8]. The absence of aniridia does not exclude the diagnosis [3]. The other ocular manifestations reported in the registry were nystagmus (93.9%), cataract (86.1%), glaucoma (55.7%), myopia (48.3%), optic nerve hypoplasia (40.7%), foveal or macular hypoplasia (37.9%), amblyopia (28.4%), strabismus (26.5%), keratopathy or pannus (24.3%), Peters anomaly (19.4%), retinal detachment (10.8%), aphakia (10.5%) and coloboma (10.1%) [8]. AAK is earlier and more severe than in aniridia due to an intragenic variant, with advanced grades already in childhood [24]. The French PNDS considers the ocular complications of WAGR comparable to those of severe forms of aniridia and cites a mean visual acuity of 1/10 to 2/10 in young adults with aniridia [7].

Wilms tumour and nephrogenic rests. The risk of Wilms tumour is 45–60% according to GeneReviews and approximately 55% in the European review of four cohorts; Smith gives 50%, and Oxford more than 50% [11, 18, 33, 34]. In the registry, 42 of 77 (54.5%) developed Wilms tumour and/or nephrogenic rests (36 tumours; 6 rests only); more than half were diagnosed by 18 months, 83.3% by 3 years and 95.2% by 5 years, with a median of 19 months [8]. In the National Wilms Tumor Study Group (NWTSG; 64 patients with WAGR among 8,533 patients with Wilms tumour, 1969–2002), the median age at diagnosis was 22 months versus 39 months in the others, 17% were bilateral (6% in the others), 2% metastatic (13%), 100% of favourable histology (92%) and 77% had intralobar nephrogenic rests (22%) [35]. In the 30-year experience of the SIOP-RTSG (43 patients), the median age was 22 months (6–44), 37.5% had bilateral disease (including nephroblastomatosis), none had metastases or anaplasia, 52.8% of tumours were of stromal type and 78.9% had nephrogenic rests [13]. Tumours have been described after 7–8 years, 25 years being the latest age reported, and metachronous contralateral tumours up to 7 years after the first (at 9 years of age) or, in the registry, a third tumour at 19 years and 7 months [3, 8, 13]. The Japanese guideline, in a synthesis of unselected aniridia series, found Wilms tumour in 0–26.9% of patients [10]. The French PNDS estimates the risk at 45–60% overall and at 40–50% in simple cases of aniridia with an 11p13 deletion including WT1; it describes tumours often in the first year of life, 90% before 4 years and 98% before 7 among those who develop a tumour, bilateral disease in more than 30% (versus approximately 5% in non-syndromic tumours), frequent nephroblastomatosis and usually favourable histology, and reports patients with deletion of PAX6 and WT1 who never had a tumour [7].

Genital and nephro-urological anomalies. In the registry, 59.2% had genitourinary anomalies (84.8% of boys and 39.5% of girls): cryptorchidism (67.7% of boys), hypospadias (35.5%), micropenis (25.0%), ambiguous genitalia (10.8% of the total), streak ovaries (35.5% of girls), bicornuate uterus (14.7%) and polycystic ovary syndrome (7.7%) [8]. Differences of sex development with a 46,XY karyotype and female external genitalia have been described, and gonadoblastoma is occasional; it was not recorded in the SIOP-RTSG series, and the risk appears lower than that of intragenic WT1 variants, especially of the Frasier type [3, 13, 18]. Congenital anomalies of the kidney and urinary tract (CAKUT) affected 38.5% (ectopic, duplex or horseshoe kidney, unilateral agenesis, ureteral duplication, hydronephrosis), recurrent urinary tract infections 20.5% and difficulty in bladder emptying 10% [3, 8]. In girls, the external genitalia are usually normal and uterine or ovarian anomalies are identified only by targeted investigation, such as the pelvic ultrasound recommended by the French PNDS [7].

Chronic kidney disease. Chronic kidney disease (CKD) is multifactorial: in the registry, 33.3% reported proteinuria, 25.0% renal failure and 19.4% focal segmental glomerulosclerosis, and 24% of those with renal failure or glomerulosclerosis had not had Wilms tumour [8]. In the IWSA survey, CKD occurred in 5 of 81 children with WAGR without Wilms tumour (6.2%) and in 34 of the 64 with Wilms tumour; in multivariate analysis, prior nephrectomy was the only significant factor [36]. Historical estimates indicated 38–53% chronic renal failure 20 years after diagnosis of the tumour, and Oxford gives approximately 40% at 20 years; Smith gives 50% renal failure [8, 18, 34]. Obesity, hypertension and the cardiometabolic profile are associated with the severity of CKD [3, 8]. The French PNDS distinguishes CKD related to the tumour (nephron reduction, chemotherapy toxicity) from that which arises independently, through focal segmental glomerulosclerosis lesions, whose survival is similar to that of non-syndromic forms and depends on access to kidney transplantation; according to the old NWTSG studies it cites, 34–40% of Wilms tumour survivors with WAGR reached end-stage kidney disease, and it estimates the risk of renal failure in children with WAGR at 50–60% [7].

Neurodevelopment and behaviour. Approximately 77% of registry participants reported cognitive impairment and/or global delay, 62.8% expressive language delay, 32.1% receptive delay, 21.8% auditory processing disorder, 44.1% anxiety disorder, 29.9% obsessive-compulsive disorder, 25.0% ASD, 23.7% ADHD and 19.0% behavioural disorder [8]. Smith describes moderate to severe intellectual disability in most, microcephaly and growth deficiency in at least 50% [18]. In the SIOP-RTSG series, 18 of 22 patients with information had cognitive impairment [13]. The French PNDS describes intellectual disability (intelligence quotient below 74) in 70%, noting that some people with WAGR have normal intelligence without behavioural changes, and abnormalities on neurological examination in up to 30%: peripheral hypertonia and/or axial hypotonia, epilepsy and microcephaly, sometimes with central auditory deficits [7]. Brain imaging may show, without specificity for WAGR, ventricular dilatation, agenesis of the corpus callosum, periventricular heterotopias or cerebellar hypoplasia, and the PNDS illustrates a bilateral cerebellar hyperintensity on MRI; in a child with marked obesity, a more extensive deletion (WAGRO) should be considered [7].

Other systems. Obesity occurred in 52.7% (usual onset by age 10), short stature in 47.8%, hypertension in 35.4% and hyperlipidaemia in 25.4%; 11.8% met criteria for possible metabolic syndrome [8]. Respiratory infections, obstructive sleep apnoea (30.1%), asthma (25.4%), chronic constipation (52.6%), feeding difficulties (48.1%), gastro-oesophageal reflux (36.0%), hypotonia (55.9%), scoliosis (19.7%) and seizures (18.1%) are frequent; pancreatitis was reported by 7.2% [8]. Structural congenital heart defects were uncommon and showed no pattern (atrial or ventricular septal defect, patent foramen ovale; two cases of tetralogy of Fallot) [3]. Hearing loss requiring hearing aids affects 5–10% [3]. Deletions involving BDNF are associated with reduced pain sensitivity, with an attenuated response to acute stimuli [3]. Tracheomalacia was reported by 13% [3]. Craniofacial features are non-specific: low-set, malformed ears, downslanting palpebral fissures, micrognathia and prominent lips [3, 18]. In Fischbach's series of 54 cases, reproduced by Hall and colleagues, non-classical manifestations included cryptorchidism (61.3%), recurrent sinusitis (27.8%), proteinuria (25.9%), obstructive sleep apnoea (20.4%), ASD (18.5%), obesity (18.5%), optic nerve hypoplasia (14.8%), focal segmental glomerulosclerosis (11.1%), epilepsy (7.4%) and chronic pancreatitis (5.6%) [2].

4. Clinical Recognition and Warning Signs

Reasons for consultation in the first months of life are nystagmus, photophobia, dark eye colour (particularly in families with light irises), wandering gaze, buphthalmos (a sign of congenital glaucoma), corneal opacity, leucocoria (congenital cataract) and microphthalmia [9]. A 2023 review summarises four typical ocular presentations: “black” eyes at birth, nystagmus in infancy, low vision due to foveal hypoplasia and childhood cataract; aniridia may be confused with bilateral congenital mydriasis [24]. Photophobia is a historical feature guiding the differential diagnosis of nystagmus, alongside achromatopsia and albinism [37].

The warning sign with the greatest consequences is aniridia in an infant without a family history. In this situation, the first step is to exclude an 11p13 deletion including WT1, using CMA, and the child should be considered at risk of Wilms tumour, with renal ultrasound every 3 months while investigations proceed [1, 8, 11]. Oxford recommends examining the parents of sporadic cases and notes that, in a cohort of 77 patients with sporadic and familial aniridia studied using an 11p13 cosmid panel, approximately 40% had chromosomal rearrangements [27]. The probability of a deletion involving WT1 is very low in familial aniridia, with only two families with WAGR described [2].

Warning signs of WAGR, besides aniridia, are: hypospadias, cryptorchidism or ambiguous genitalia; developmental delay; microcephaly and growth deficiency; Wilms tumour with aniridia; and, in a child with Wilms tumour, predominantly stromal histology and intralobar nephrogenic rests [18, 27, 34, 38]. In the evaluation of a child with Wilms tumour, examination of the eyes for aniridia, blood pressure and urinalysis for proteinuria are part of the search for an associated syndrome [33, 34]. In an SIOP-RTSG series, 30.8% of tumours in patients with WAGR presented as a palpable mass or with symptoms, and in two the diagnosis of WAGR was made only after the tumour [13]. The French PNDS notes that, without screening, the tumour usually presents as an abdominal mass, and that haematuria, malaise (due to hypertension, present in approximately 25%) or pain (due to fissuring or rupture) occur in 20–30% of children with Wilms tumour [7].

The same PNDS assigns specific tasks to the general practitioner: to remember that aniridia should be recognised at birth or at child health surveillance visits from 2 months of age; when aniridia is present, to look for Wilms tumour, urogenital anomalies and developmental delay and to request testing for the 11p13 deletion; to look for other manifestations (cognitive difficulties, diabetes, balance disorders, deafness, changes in sense of smell); to liaise with ophthalmology, nephrology and neurology; and to direct the family to psychological support and social entitlements [7]. When a urogenital anomaly is present, from isolated cryptorchidism to a more severe difference of sex development, the PNDS requires a search for aniridia [7].

Pitfalls can delay diagnosis. In mild forms, the iris may appear normal and only foveal hypoplasia on macular OCT reveals the disease, so OCT should be performed when iris defects are subtle [1, 24]. Apparently healthy parents may have mild iris hypoplasia or reduced acuity due to foveal hypoplasia detectable only on specialist examination [1, 27]. The expression “sporadic aniridia” is misleading because it suggests no recurrence risk, and GeneReviews prefers “simplex case” [1]. An iris with a scalloped margin and central aplasia to the collarette, with fixed mydriasis, points towards Gillespie syndrome or ACTA2-related multisystemic smooth muscle dysfunction rather than PAX6 aniridia [2]. Cerebellar ataxia with aniridia suggests Gillespie syndrome and warrants brain MRI [9, 39].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

GeneReviews states that there are no consensus clinical diagnostic criteria for PAX6 aniridia syndrome or for WAGR spectrum disorder; the diagnosis is established in a proband with suggestive findings and a proven genetic alteration [1, 3]. The suggestive findings are complete or partial iris hypoplasia on slit lamp, with transillumination, abnormal architecture or pupillary changes, accompanied by reduced acuity due to foveal hypoplasia or optic nerve anomalies, progressive keratopathy, cataract and glaucoma, early nystagmus, microphthalmia or coloboma, and a family history compatible with dominant inheritance, although its absence does not exclude the diagnosis [1]. For WAGR, the suggestive findings are Wilms tumour or nephrogenic rests, aniridia or other eye disease, genital anomalies, differences of sex development, CAKUT, developmental delay and neurobehavioural problems [3]. The historical clinical criterion for WAGR syndrome required aniridia and at least one of the other three features, which classified 90% of registry participants; the spectrum criteria classified 95–100% [8]. The French PNDS adopts a broader rule: the diagnosis requires at least two of the four components (risk of Wilms tumour, total or partial aniridia with the associated panocular anomalies, urogenital anomalies, neurodevelopmental disorder) and/or an 11p13 deletion, and states that no recognised classification exists [7].

The Japanese guideline published formal criteria. Symptoms (A) are bilateral visual impairment and photophobia; findings (B) are iris hypoplasia (B1), macular hypoplasia on fundus examination or OCT (B2), limbal corneal lesions (B3), cataract (B4), microphthalmia (B5), nystagmus (B6) and glaucoma (B7); exclusions (C) are post-herpetic iris atrophy, traumatic or surgical defects, coloboma due to failed closure of the optic fissure, Rieger anomaly and iridocorneal endothelial syndrome; criteria also include extraocular complications (D), genetic testing (E, a pathogenic PAX6 variant or 11p13 deletion) and affected relatives (F) [10]. Diagnosis is “definite” with any item from A, B1 and E and exclusion of C; “probable” with A, B1 and F, or A, B1 and B2, or A, B1 and B3; and “possible” with A and B1 without complete exclusion of C [10]. The same guideline defines severity in four grades, from grade I (one affected eye and the other healthy) to grade IV (both affected and corrected acuity in the better eye below 0.1), with an increase of one grade when secondary glaucoma narrows the visual field of the better eye [10].

5.2. Ophthalmological evaluation and imaging

The European guidelines recommend, in a child with suspected aniridia, a complete ophthalmological examination under general anaesthesia at the first assessment, with inspection of the face and grading of ptosis, visual acuity, cycloplegic refraction and orthoptic assessment of nystagmus and strabismus, followed by slit-lamp examination and anterior segment photography to document the cornea, the full 360° of the limbus, iris and lens [9]. AAK is graded on a scale from 0 to 4 according to invasion of the limbal margin and visual axis, as shown in the table [9].

Source: table 1 of the ANIRIDIA-NET guidelines, adapted from Lagali and colleagues; representative images of grades 0 to 4 appear in figure 1 of the article [9].

GeneReviews recommends routine OCT to confirm the diagnosis, especially when iris defects are subtle; AS-OCT delineates anterior segment anatomy even with an opaque cornea, macular OCT documents and grades foveal hypoplasia, B-mode ultrasound establishes baseline axial length and may suggest optic nerve hypoplasia, and high-frequency UBM, usually under anaesthesia, identifies iris hypoplasia in infants with an opaque or oedematous cornea [1]. Spectral-domain OCT has prognostic value for visual acuity, although with small samples [1]. The European guidelines add that OCT, topography and confocal microscopy may be difficult or inconsistent because of nystagmus and AAK, that portable devices help in children, and that visual fields are unreliable [9]. IOP should be measured with gonioscopy, usually every 6 months, and corrected for pachymetry [9]. The French PNDS, for the young child with aniridia in the context of WAGR, lists the content of the initial examination under general anaesthesia, as a day case or inpatient: objective refraction, IOP, corneal measurements, slit lamp, examination of the angle and lens, fundus, pachymetry, axial length, UBM, anterior and posterior segment OCT and fundus photography, possibly wide-field, acknowledging that not all are feasible with the equipment available; examinations under anaesthesia, initially annual, are spaced out according to the course and the child's cooperation [7].

The initial assessment recommended by GeneReviews also includes the history of visual behaviour observed by parents, best-corrected acuity (with support from orthoptists and acuity cards in infants), refraction, indirect ophthalmoscopy, fundus autofluorescence and baseline fundus photography; beyond the eye, developmental and neurobehavioural assessment, brain MRI, investigation of central auditory difficulties and sleep disorders, endocrine assessment for obesity and glucose intolerance, and genetic counselling [1]. In children, the European guidelines recommend a complete paediatric assessment to exclude WAGR, renal ultrasound while genetic results are awaited and, if developmental delay is present, neurological examination and MRI to exclude cerebellar anomalies of Gillespie syndrome [9].

5.3. Genetic tests: strategy, yield and limitations

GeneReviews and the European guidelines recommend genetic testing in all suspected cases [1, 9]. In an infant with aniridia without previous testing, the first investigation should be CMA to detect 11p13 deletions including PAX6 and WT1, given the implications for Wilms tumour screening [1, 33]. The strategy proposed by Hall and colleagues combines two tests in almost all cases: a next-generation sequencing (NGS) panel including at least PAX6, FOXC1, PITX2 and ITPR1, and copy-number analysis, such as comparative genomic hybridisation array (array-CGH); these authors do not distinguish investigations of simplex and familial cases, except when the familial variant is already known [2]. Single-gene testing (PAX6 sequencing followed by deletion analysis) is rarely useful and generally not recommended; a panel may be too restricted in the setting of complex deletions and structural variants [1]. The Japanese guideline considers counselling and testing of PAX6 and neighbouring genes by experts useful and notes that test choice depends on insurance coverage [10].

The proportion of variants detected by each method does not equate to the clinical sensitivity of the test: it depends on the population studied, pre-test probability and platform coverage, and a negative result does not exclude diagnosis in the presence of a convincing phenotype [1]. The proportion of classical aniridia without a genetic explanation has been estimated at approximately 5% [2]. In the Japanese guideline's systematic review, sequencing identified variants in nearly 85% of isolated aniridia cases, and MLPA or CMA identified alterations of PAX6 or adjacent regions in nearly 15% [10]. In 2026, a French genome sequencing study in 100 families with previously undiagnosed developmental eye malformations identified pathogenic or likely pathogenic variants in 18 and variants of uncertain significance in 9, 14 of which would not have been detected by panels or array-CGH [41]. A PAX6 variant of uncertain significance neither establishes nor excludes the diagnosis [1].

In sporadic aniridia in a child without a family context, the French PNDS recommends CMA as the first test, to confirm the 11p13 deletion and measure its size; as alternatives or complements, FISH with probes for PAX6 and/or WT1 and for the PAX6 regulatory region in ELP4 (which may be deleted with the PAX6 locus intact), MLPA or high-resolution karyotype (at least 550 bands per haploid set), the latter now little used but useful, with FISH, for characterising parental rearrangements [7]. Conventional karyotyping detects balanced parental rearrangements involving segments larger than 5–10 Mb [7]. The PNDS illustrates the CMA result in a child with WAGR, arr(hg19) 11p14.3p12(25,791,559-42,648,146)x1, a deletion of approximately 16.9 Mb, and that of an interphase FISH with a single 11p13 signal [7].

The report must specify the gene, transcript with version (the MANE Select for PAX6 is NM_001368894.2; for WT1 it is NM_024426.6), complementary DNA and protein HGVS nomenclature, zygosity, classification according to the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP), and method limitations; when the variant is described in the literature using NM_000280 numbering, both should be included [1, 21]. For WAGR, the report must define deletion size and list the genes involved (WT1, BDNF, ELP4, ALX4, EXT2), because this changes surveillance [3, 6]. Short-read WGS detects copy-number variants more reliably than exome sequencing, but each region must be checked for coverage and quality, and WAGR deletions require robust copy-number variant (CNV) calling algorithms [1, 3].

5.4. Practical algorithm

The following algorithm integrates recommendations from GeneReviews, the European guidelines, the review of Hall and colleagues and the WAGR screening programme [1, 2, 8, 9].

The discontinuation of renal screening when WT1 is intact follows from the fact that deletions and intragenic variants confined to PAX6 do not confer a risk of Wilms tumour; the Japanese guideline formulates this in the same way when sequencing identifies the pathogenic variant in PAX6 without a structural alteration [5, 10, 13]. The decision tree of the French PNDS starts from two entry points, aniridia and genitourinary anomaly (the latter leads to a search for aniridia), and, once the deletion is confirmed, directs to nephrological follow-up with renal ultrasound every 3 months until 6–8 years, without regular monitoring after 8 years in the absence of a tumour, neurological follow-up and a search for obesity (WAGRO) and for the bone anomalies of more extensive deletions [7].

5.5. Prenatal and preimplantation diagnosis

Once the familial genetic alteration has been identified, prenatal diagnosis and preimplantation genetic testing are possible [1, 3]. Second-trimester ultrasound visualises the eye and lens, but is unlikely to be sensitive enough to detect recurrence [27]. There are no specific prenatal findings of WAGR, although oligohydramnios has been described in the prenatal history of children diagnosed later [3]. Oxford notes that prenatal diagnosis of aniridia is rarely requested and that it does not inform on clinical severity [27]. In WAGR, the French PNDS reserves prenatal diagnosis mainly for couples in whom a parent has a balanced chromosomal rearrangement involving 11p13: chorionic villus sampling at 12 weeks of gestation or amniocentesis from 15 weeks is offered, with CMA, karyotype and/or FISH according to the context, and preimplantation diagnosis is possible if the couple is eligible for in vitro fertilisation with intracytoplasmic sperm injection, which avoids a termination of pregnancy for medical reasons [7]. Without a parental rearrangement, the risk of recurrence is low, but germline mosaicism cannot be excluded, so the PNDS requires an invasive sampling to be discussed in subsequent pregnancies; the FISH results in the proband serve as the basis for these tests [7].

6. Differential Diagnosis

The differential diagnosis is organised along three axes: other genetic causes of aniridia or iris hypoplasia, other anterior segment dysgeneses, and other causes of Wilms tumour with genitourinary anomalies [1, 27, 34].

In a child with aniridia and ataxia, Oxford and the European guidelines suggest Gillespie syndrome and recommend brain MRI [9, 39]. Two cases with a PAX6 variant and Gillespie features were described before identification of ITPR1, but without the scalloped iris appearance [2]. In anterior segment anomalies, allelic and genetic heterogeneity is the rule: truncating PAX6 variants cause aniridia and paired-domain missense variants cause some cases of Peters anomaly, whereas the same phenotype may result from PAX6, MAF, PITX2 or CYP1B1 [27]. Foveal hypoplasia on OCT favours PAX6 aniridia over Gillespie syndrome, in which it is typically absent [1, 2].

7. Treatments and Risk Management Strategies

7.1. Principles and organisation of care

There is no curative treatment or approved targeted therapy for aniridia or for WAGR; treatment is supportive, multidisciplinary and aimed at preserving vision, reducing symptoms and detecting complications early [1, 3, 9]. GeneReviews bases its recommendations on the authors' experience, the European guidelines on a literature review without formal grading and the Japanese guideline on systematic reviews with strength of recommendation and level of evidence; in all of them, the evidence comes from case series, with no randomised trials for any management strategy [1, 9, 10]. The European guidelines recommend referral to a reference centre, led by the ophthalmologist, with involvement of other professionals according to the manifestations and genotype, and stress that, for complex procedures, improvement of visual acuity should not be the main objective, but rather symptoms and quality of life [9]. A scoping review published in August 2026 found mainly observational evidence and expert recommendations on organisational models, and no comparative evaluation between models [46]. In WAGR, the French PNDS (French National Diagnosis and Care Protocol) organises care around the reference or competence centre, with ophthalmologist, paediatrician, cytogeneticist, clinical and molecular geneticists and nephrologist in all cases and, depending on the picture, paediatric neurologist, oncologist, paediatric surgeon, urologist, radiologist, gynaecologist, endocrinologist, otorhinolaryngologist, psychiatrist and rehabilitation physician; among non-medical professionals it includes orthoptists, opticians specialised in low vision and in children, psychomotor therapists, specialised educators, speech and language therapists and occupational therapists, physiotherapists, psychologists, social workers, dietitians, hearing-aid specialists and independence instructors for people with visual impairment, a profession created in France in May 2020 [7]. The centre alternates follow-up with the local ophthalmologist and the attending physician, and enrolment in a care programme should begin as soon as visual impairment or developmental delay is detected [7].

7.2. Refraction, photophobia, amblyopia and low vision

Correction of refractive errors is the first preventive measure; spectacles are preferred, with tinted or photochromic lenses for the photophobia associated with the large pupillary aperture [1, 9]. Anisometropic or strabismic amblyopia is treated with occlusion or optical penalisation, taking nystagmus into account; strabismus surgery improves alignment and appearance, but not visual function [1, 9]. Optical low-vision aids (magnifiers, telescopes, video magnifiers and closed-circuit television systems) and non-optical aids (large-print books, adequate lighting, filter lenses) benefit children and adults, and orientation and mobility and rehabilitation specialists contribute to independence [1, 9]. For outdoor activities, the European guidelines recommend photochromic or category 4 lenses with UVA/UVB protection [9]. The French PNDS adds sunglasses and swimming goggles with 100% UVA and UVB protection, also to protect the cornea from salt water, chlorine, wind, dust and snow, tinted car windows, a wide-brimmed hat or cap, a lux meter to measure light intensity, monoculars or binoculars to read distant information, a white cane (including electronic), a guide dog and voice navigation systems, with specific mobility training [7]. In amblyopia, it accepts occlusion or optical penalisation with Ryser lenses, always weighed against low vision and nystagmus [7].

There is divergence regarding contact lenses. The European guidelines strongly advise against soft contact lenses and any coloured, filter or prosthetic contact lens, because of surface hypoxia that may precipitate or accelerate limbal deficiency, and accept scleral lenses, which do not touch the cornea [9]. GeneReviews accepts coloured, filter or artificial-pupil lenses, warning that they may be poorly tolerated because of the ocular surface and tear film [1]. The Japanese guideline, by contrast, makes a strong recommendation (evidence C) for filter lenses and contact lenses with an artificial iris («Iris Lens») for photophobia and for low-vision rehabilitation [10]. The French PNDS aligns with the European guidelines: it strongly advises against soft contact lenses and coloured therapeutic lenses, because of keratopathy, tear film changes and dry eye, and prefers tinted or photochromic spectacles [7]. The choice should weigh the state of the ocular surface and the risk of worsening AAK, and the difference reflects different jurisdictions and practices.

7.3. Ocular surface and keratopathy

The European guidelines recommend preservative-free lubricants as soon as the first AAK signs appear, frequently in infants and young children; in more advanced dry eye disease, ciclosporin 0.1% or autologous or allogeneic serum eye drops may be added, and scleral lenses protect the cornea while keeping it lubricated [9]. GeneReviews adds mucolytics and punctal occlusion and notes that preservative-free eye drops avoid additional toxicity [1]. In the European Union, ciclosporin 1 mg/ml (Ikervis®) has been authorised since 19 March 2015 only for severe keratitis in adults with dry eye disease that has not improved with tear substitutes, at a dose of one drop in the affected eye at bedtime, with reassessment of the need to continue treatment at least every 6 months; use in paediatric aniridia is off-label [47].

Surgical treatment of AAK carries a high risk of failure because of chronic inflammation, neovascularisation and limbal deficiency [9]. Limbal stem cell transplantation cannot be autologous because the disease is bilateral, and requires permanent systemic immunosuppression; keratolimbal allografting and living-related donor transplantation have 5-year success rates between 25% and 81.4%, without controlled studies and with heterogeneous definitions of success, and even cultured HLA-matched transplantation has generally produced poor results [9]. Penetrating keratoplasty, with or after limbal transplantation, has 5-year graft survival of 43–50%, and type 1 Boston keratoprosthesis retention of approximately 80% at 3–5 years, but anatomical success rarely translates into meaningful visual gain and entails risks of rejection, worsening glaucoma, retroprosthetic membrane, endophthalmitis and extrusion [9]. An individual-patient-data meta-analysis published in 2026 (26 series, 325 eyes) found anatomical success of 82% with keratoprosthesis versus 38% penetrating graft survival (failure hazard ratio 2.92; 95% CI, 1.24–6.84), with no difference in acuity gain or complications [48]. In Jacobson's North American series, keratoprosthesis had the longest survival at 10 years (64%; 95% CI, 32–84), but penetrating keratoplasty and keratoprosthesis were associated with worse final acuity [30].

The Japanese guideline recommends, with weak strength, not performing keratoplasty for stromal opacity (evidence B), because of limited functional benefit and poor long-term prognosis, and recommends, with weak strength (evidence C), surgical treatment of limbal deficiency with allogeneic limbal transplantation or cultivated oral mucosal epithelial transplantation, combined with keratoplasty when there is stromal opacity; in the interim results of 17 eyes treated with autologous oral epithelial sheets, the cornea became clear in 76.4% and acuity improved in 88.2% [10]. GeneReviews accepts superficial keratectomy for opacification with visual impact and considers the prognosis of isolated penetrating keratoplasty poor [1]. The French PNDS is the most restrictive: it considers keratoplasty contraindicated when there is limbal deficiency, recommends active monitoring and treatment of dry eye and corneal transparency, and requires eye drops and gels to be preservative-free, in single-dose units or dedicated bottles [7].

7.4. Glaucoma

Initial treatment is medical, with gradual introduction and maintenance for as long as possible to delay surgery; beta-blockers, prostaglandin analogues and carbonic anhydrase inhibitors may be used, preferably preservative-free [1, 9]. By extrapolation from the general paediatric population, the European guidelines indicate beta-blockers as first line below 9 years and prostaglandins from 9 years [9]. Brimonidine requires caution: the European guidelines consider it contraindicated below 8 years and to be avoided until 12, because of somnolence and lethargy (76%), ocular pruritus (49%) and burning (39%) in a paediatric series [9]. The US label of Alphagan P® (brimonidine 0.1% and 0.15%) contraindicates it below 2 years, describes apnoea, bradycardia, coma, hypotension, hypothermia, hypotonia, lethargy and respiratory depression in infants, and reports somnolence in 50–83% of children aged 2 to 6 years treated with brimonidine 0.2% three times daily [49]. The European restriction is therefore broader than that of the label. The French PNDS requires IOP measurement at the initial assessment and then every 6 months, depending on age, and treatment of glaucoma, as far as possible, with preservative-free eye drops [7].

Sources: [1, 9, 10]. A meta-analysis of 22 studies published in 2025 found no difference in complete success between trabeculectomy and drainage devices, but qualified success was greater with devices, with higher failure risk after trabeculectomy (hazard ratio 5.68; 95% CI, 1.46–22.14), without differences in IOP reduction, medication reduction or acuity [50]. In Jacobson's series, half of eyes with glaucoma required surgery, and drainage devices had 71% success [30]. The Japanese guideline recommends inserting the tube tangentially, avoiding contact with the endothelium and lens in the absence of an iris [10].

7.5. Cataract

Surgery should be delayed until opacity is severe, because foveal hypoplasia limits visual gain and because surgery in aniridia may precipitate or worsen glaucoma, accelerate AAK and cause endothelial decompensation [1, 9]. A longitudinal comparison between operated and unoperated eyes showed no significant improvement in mean acuity [1]. The Japanese guideline weakly recommends (evidence D) surgery when indicated, informing the patient beforehand of the risks of worsening glaucoma, AFS and bullous keratopathy [10]. Technically, zonular and anterior capsule fragility favour capsular staining, cohesive viscoelastic devices and a capsulorhexis diameter balancing rupture risk against fibrosis and possible implantation of an iris prosthesis in the capsular bag [9]. Black-diaphragm aniridic intraocular lenses reduce glare but are associated with progression of glaucoma and AAK, cystoid macular oedema, chronic endothelial loss and explantation [1].

7.6. Iris prostheses

Iris prostheses include the iris–lens diaphragm, capsular tension ring devices and the customised artificial iris [9]. The CustomFlex® artificial iris (HumanOptics) was marketed as a custom-made device from 2002, received CE marking in 2011 and was approved by the Food and Drug Administration (FDA) through a premarket approval application (PMA P170039) on 30 May 2018, for children and adults with complete or partial congenital, acquired or other-condition-associated aniridia [9, 51]. The foldable acrylic iris–lens diaphragm has CE marking but no FDA approval, and the BrightOcular artificial iris has neither and has been associated, in healthy eyes, with uveitis, glaucoma and corneal decompensation [9]. In a systematic review, prostheses reduced photophobia in 96% (CustomFlex) and 90% (diaphragm) and were associated with acuity gain in 62.5%, generally 2–3 logMAR lines, probably attributable to simultaneous cataract surgery; complications included progression of pre-existing glaucoma (58.9%), secondary glaucoma (27.6%), AAK progression (27%), decentration (10.6%) and endothelial loss (9%, with diaphragm only) [9]. The European guidelines conclude that prostheses should not be offered as first-line surgery [9].

7.7. Aniridic fibrosis syndrome

Surgical intervention is recommended at the first sign of AFS, and patients with multiple previous operations should be monitored for this complication during the annual examination [1]. Before any intraocular operation, the patient should be informed of the risk of triggering or worsening AAK and AFS [1].

7.8. Development, behaviour and systemic problems in PAX6 aniridia

GeneReviews recommends early intervention from ages 0 to 3, individualised educational programmes with support from vision specialists, speech and language therapy with assessment for augmentative and alternative communication when indicated, management of behaviour by developmental paediatrics and, when appropriate, interventions used in ASD and medication for ADHD; in the United Kingdom, the education, health and care plan (EHC plan) is reviewed every 6 to 12 months [1]. Given central auditory processing difficulties, preservation of hearing and adaptation of the school environment are relevant [1]. A review recommends a low threshold for assessing diabetes in patients with a PAX6 variant, although glucose intolerance is rare in childhood and the best screening strategy is undefined [24].

7.9. WAGR spectrum disorder: Wilms tumour surveillance

The aim of surveillance is to detect tumours of smaller volume and lower stage, requiring less treatment and allowing nephron-sparing surgery, particularly important in a population at risk of renal failure [11]. Tumours detected through surveillance had a median volume of 18 ml versus 375 ml in symptomatic cases (p=0.001), and no studies directly compare survival between screened and unscreened children [11, 13]. Renal ultrasound is the recommended method, without radiation or anaesthesia, using high-frequency probes and assessing the entire parenchyma in supine and prone positions; in WAGR, examination of the kidneys suffices, unlike Beckwith–Wiedemann syndrome, which requires complete abdominal ultrasound [11]. Surveillance should be offered after risk counselling by a geneticist or genetic counsellor, start at birth or at clinical suspicion without awaiting molecular confirmation, and follow-up should be coordinated by cancer predisposition specialists [8, 11].

Differences between recommendations arise from retrospective data and different risk criteria (1% in North America, 5% in Europe) and duration criteria (age by which 90–95% of tumours occurred); none has been evaluated in randomised studies [11, 12, 34]. Surveillance does not replace assessment of symptoms such as an abdominal mass or haematuria between examinations [13].

7.10. Wilms tumour treatment and renal protection in WAGR

Patients with WAGR and Wilms tumour can be treated according to existing protocols by experienced multidisciplinary teams, with preoperative chemotherapy to facilitate nephron-sparing surgery, recommended for children with genetic predisposition whenever tumour size and location permit [11, 13]. In the SIOP-RTSG series, 92.9% received preoperative chemotherapy (actinomycin D and vincristine in most); volume decreased in 50% and increased in 42.9%, particularly in stromal tumours and nephroblastomatosis; 85% of bilateral and 23.8% of unilateral cases underwent nephron-sparing surgery; two children died from hepatic sinusoidal obstruction syndrome during preoperative chemotherapy, one after an actinomycin D overdose, leading the authors to recommend intensive toxicity monitoring [13]. In the IWSA survey, complete nephrectomy was the only independent factor associated with CKD [36].

GeneReviews, adapting the WAGR registry, structures lifelong renal surveillance: annual measurement of blood pressure against age-specific tables, urinary albumin/creatinine ratio and serum creatinine; closer surveillance after Wilms tumour, nephrectomy or low birth weight (<2.6 kg, approximately 26%); education on nephron-protective measures, such as avoiding non-steroidal anti-inflammatory drugs; a low threshold for urine culture in febrile illnesses; monitoring of glycaemia and lipids; baseline renal ultrasound for CAKUT, size and echogenicity, repeated as determined by the nephrologist; and transition to adult nephrology without relaxing surveillance in adolescence, because CKD may arise 2 to 13 years after treatment of the tumour and the risk of end-stage kidney disease accumulates after 12 years [3]. Oxford recommends, in WT1 syndromes without nephropathy, annual blood pressure and urinalysis because of the risk of renal failure in adulthood [34]. The French PNDS provides for blood pressure and urine sediment every 6 months, to detect the proteinuria of a non-tumoural glomerulopathy, serum creatinine whenever in doubt, assessment of renal function throughout life (blood pressure, proteinuria, creatinine-based estimated glomerular filtration rate), especially after bilateral tumour, nephrological follow-up of all children treated for kidney cancer and follow-up in paediatric oncology until adulthood to detect relapse, contralateral tumour and sequelae [7]. For treatment of the tumour, it describes the European regimen of the UMBRELLA protocol, with neoadjuvant chemotherapy and nephron-sparing surgery whenever possible in syndromic forms; the details of the regimen, stages and radiotherapy are in the chapter on predisposition to Wilms tumour [7].

The French PNDS diverges from the other recommendations on several points, which should be made explicit. In tumour surveillance, the 3–4-month interval is wider than the 3 months of SIOP-Europe, AACR and the registry, and the duration (7 years in the text, 6–8 years in the decision tree, without regular monitoring after 8) is shorter than that of the registry and GeneReviews, which after 8 years maintain shared decision-making and renal ultrasound at least annually throughout life [3, 7, 8, 11, 12]. In renal surveillance, it requires six-monthly blood pressure and urine, versus the annual interval of GeneReviews [3, 7]. While awaiting the genetic result, it accepts ultrasound every 1 to every 3 months, versus the 3 months of the other sources [7, 11]. On these points, the European SIOP recommendations and the registry programme, which are more recent and, in the case of SIOP, have explicit grading of evidence, serve as the reference, and the PNDS as comparison.

7.11. Other measures in WAGR

The initial assessment includes complete abdominal and pelvic ultrasound, examination by an ophthalmologist experienced in aniridia, by urology and gynaecology (internal and external genital anomalies, differences of sex development), nephrology, neurology (MRI if indicated), development, orthopaedics and physical medicine (scoliosis), neuropsychiatric assessment from 12 months (sleep, ADHD, anxiety, ASD), gastroenterology and nutrition (aspiration risk), audiology, assessment for obstructive sleep apnoea and genetic counselling [3, 8]. Ophthalmological measures include frequent preservative-free artificial tears, avoidance of contact lenses except bandage lenses, sunglasses and hats [3]. Obesity requires a multidisciplinary approach, with regular physical activity and nutritional intervention if BDNF is deleted or there is hyperphagia, similar to the recommendations for Prader-Willi syndrome [3]. Anaesthetists should be alerted to the possibility of tracheomalacia, respiratory difficulty after general anaesthesia and velopharyngeal insufficiency [3]. In the reduced pain sensitivity associated with BDNF deletion, objective signs of injury or illness (swelling, limping, reduced activity) should be monitored rather than facial expressions or verbal complaints [3]. Preservation of hearing, with timely treatment of otitis, is a priority in a population with impaired vision [8]. The French PNDS includes in the initial assessment blood and urine tests, pelvic ultrasound for genital malformations, neurological examination with brain imaging if necessary, paediatric neurological assessment with MRI, paediatric endocrinological assessment and specialised paediatric nephrological assessment, and recommends detailed hearing assessment in all children with WAGR [7].

For intellectual disability, the PNDS advocates early intervention by a team that includes a neuropsychologist, psychologist and speech and language therapist, in liaison with teachers, educators and school support staff, and sometimes in a medico-social establishment; the principle is to adapt the environment to the person's abilities, to build on preserved cognitive functions and to try to re-educate or stimulate deficient functions [7]. Psychological support serves to detect behavioural disorders (anxiety, mood disorder, depression, obsessive-compulsive behaviours, psychosomatic manifestations), to prepare for surgery and, later, to live with a progressive eye disease, in a context in which visual impairment is often invisible to others [7]. The PNDS recognises that, in WAGR, the degree of intellectual disability often determines independence in adult life and working life [7].

7.12. Regulatory status, clinical trials and recent developments

No medicine has an approved indication for aniridia or WAGR in the United States or European Union; the Drugs@FDA search found no ataluren records [52, 53].

Registered interventional trials are few, small and mostly completed.

These data require cautious interpretation. STAR tested a single drug in a restricted genotypic group, with few patients evaluable for the primary endpoint, heterogeneous ages and AAK grades, and a functional endpoint (reading speed) relatively insensitive to slow structural changes; absence of a significant difference does not demonstrate absence of effect, but neither supports use outside a trial [14, 24]. The European guidelines describe NCT04117880 as a registered extension, but the registry indicates that it was withdrawn before starting, and refer to ataluren eye drops as “positively tested”, when the cited study assessed only the formulation and physical and microbiological stability of 1% oily eye drops over 60 days, without patients [9, 55, 57]. Results of surgical series and meta-analyses depend on heterogeneous retrospective series and do not allow techniques to be ranked by indirect comparison [48, 50].

Developments in the last 12 months fall into distinct evidence levels. In guidance documents, the European ANIRIDIA-NET guidelines and Japanese guideline were published, and GeneReviews updated its aniridia chapter (June 2026) and created its WAGR chapter (March 2026) [1, 3, 9, 10]. In secondary clinical evidence, an individual-patient-data meta-analysis favoured keratoprosthesis over penetrating keratoplasty for anatomical success [48], and a French study documented the additional yield of genome sequencing in ocular malformations [41]. In regulatory information, PTC Therapeutics withdrew the US ataluren authorisation application for Duchenne in February 2026, closing the drug's regulatory pathway for now [54]. In preclinical research, a minimally humanised mouse model carrying the most common recurrent pathogenic variant enabled testing of an adenine base editor delivered by lipid nanoparticles, with correction of 4.8% of the variant in cortical neurons ex vivo [58]; patient-derived corneal organoids showed increased PAX6 protein with duloxetine in one patient and ataluren in another, with no response in a third [59]; previously, induced pluripotent stem cell models had identified amlexanox as the most promising inducer of premature termination codon readthrough [60]. A 2026 review described molecular targets in AAK (microRNA, inflammation, oxidative stress) [25]. None of these preclinical approaches had a registered clinical trial in the searches performed.

7.13. Reference networks and patient associations

The relevant European Reference Network is ERN-EYE, dedicated to rare eye diseases; at the consultation date, its consensus page listed no aniridia- or WAGR-specific document [15]. Patient associations include Aniridia Europe, Aniridia Network UK, Aniridia Foundation International (with a guide for new parents), the French association GENIRIS, which participated in the European guidelines, and, for WAGR, the International WAGR Syndrome Association, which maintains a patient registry on the CoRDS platform [1, 3, 9, 11]. GeneVision provides information for patients and professionals on aniridia and WAGR [1].

In France, WAGR syndrome is followed by the OPHTARA rare diseases reference centre in ophthalmology, coordinated at the Hôpital Necker-Enfants Malades (Paris), with constituent sites and competence centres in the Paris region, the other regions and the overseas departments, integrated into the SENSGENE health network (filière) for rare sensory diseases, which also brings together the CARGO, MAOLYA and REFERET centres, the keratoconus centre and the genetic deafness centre [7]. The national association GÊNIRIS supports people with aniridia and rare iris diseases, with or without an associated syndrome, and the PNDS lists other associations of people with visual impairment; internationally, it refers to ERN-EYE, EURORDIS, Aniridia Europe, Aniridia Network UK, Aniridia Foundation International, the Vision for Tomorrow Foundation and IWSA [7]. The PNDS serves as a reference for the attending physician to request exemption from patient co-payment as a condition outside the list of long-term illnesses [7].

The PNDS provides for a therapeutic education programme on aniridia, then in preparation, focused on the use of visual rehabilitation equipment and on complications (glaucoma, cataract, ptosis, nystagmus, foveal and optic nerve hypoplasia, ectopia lentis, photophobia), and recognises that intellectual disability makes this education difficult in WAGR [7]. Adaptation of lifestyle includes career guidance compatible with vision (driving is usually impossible), individualised school plans with extra time in examinations, early intervention and school support services for children with visual impairment, school inclusion units, distance learning and home teaching support, recognition of disabled worker status in adults and, for carers, respite stays, support groups and links with associations [7]. For children with bilateral visual and hearing impairment, the PNDS points to the need for a specialised school with learning of oral language and sign language [7].

8. Reproductive Options and Family Counselling

8.1. PAX6 aniridia syndrome

PAX6 aniridia syndrome is inherited in an autosomal dominant manner; approximately two-thirds of probands have an affected parent and approximately one-third have a de novo alteration [1, 9]. When the proband appears to be the only affected individual, molecular testing of both parents and ophthalmological examination for mild iris hypoplasia or reduced acuity due to foveal hypoplasia are recommended, because expressivity varies greatly within families [1, 27]. If the alteration is not found in either parent's blood, with confirmed parentage, a de novo alteration or parental gonadal (or somatic and gonadal) mosaicism should be considered; leucocyte testing does not detect all somatic mosaicism or mosaicism confined to the germline [1]. A 2023 review emphasises that late postzygotic mutations may remain restricted to a few tissues, including the eye, and escape blood analysis [24].

Sibling risk is 50% if a parent is affected or carries the alteration; if the proband's alteration is not detected in either parent's blood, risk is slightly above population risk because of gonadal mosaicism; if parents have not been tested but are clinically normal, risk appears low, but siblings are still considered at increased risk because of possible reduced penetrance or mosaicism [1]. Penetrance is almost complete, with rare reports of non-penetrance in a family with PAX6-related nystagmus [1]. Each child of an affected person has a 50% probability of inheriting the alteration, and this risk may be lower if the proband is mosaic [1]. Genetic testing does not predict severity in offspring, and Oxford notes that prenatal diagnosis is rarely requested in this condition [1, 27].

Infant children or siblings of an affected individual should have their status clarified early by ophthalmological examination or testing for the familial alteration, to initiate treatment and complication surveillance [1]. Risk determination and discussion of prenatal diagnosis and preimplantation genetic testing should occur before pregnancy, and counselling should be offered to affected young adults and parents of affected children [1]. Use of these tests is a personal decision, and differing perspectives may exist among professionals and within families [1]. Knowledge of the specific variant may also determine eligibility for future trials, as occurred with the ataluren trial restricted to nonsense variants [2, 14].

8.2. WAGR spectrum disorder

WAGR is autosomal dominant and almost always caused by a de novo 11p13 deletion [3]. It is recommended that the parents be studied with a genomic test capable of detecting the proband's deletion and with a chromosomal study to look for a predisposing balanced rearrangement, such as the reciprocal translocation t(2;11)(p21;p15) found in the father of a child with an 11p14.3p12 deletion [3]. Familial occurrence through unbalanced transmission of a balanced insertional translocation has been described, and only two families with familial WAGR have been reported [2, 18]. If neither parent has the deletion or a rearrangement, the risk to siblings is low, presumably below 1%, but higher than that of the population because of gonadal mosaicism; if a parent has a balanced rearrangement, the risk increases and depends on the rearrangement [3]. People with WAGR are not known to have had offspring, and the risk to other relatives is low, except in the case of a parental rearrangement, in which case the relatives of the parent may be carriers [3]. Prenatal or preimplantation testing may be offered even when the parents do not have the deletion, given the residual risk [3]. The French PNDS requires, as a prerequisite for counselling, karyotype and FISH (metaphases and interphase nuclei) of the parents, to exclude a balanced translocation or insertion involving 11p13 and a mosaic 11p13 deletion; it describes asymptomatic parents with the deletion in mosaic form and rare forms transmitted by affected people, usually with small deletions and a moderate phenotype, and assigns to people with WAGR a theoretical transmission risk of 50% [7]. It also recommends eye examination at birth or in early childhood of the siblings and offspring of people with aniridia, and family investigation whenever a genetic or cytogenetic alteration is identified [7]. There are two divergences from GeneReviews: the PNDS describes the risk to siblings, without a parental rearrangement, as similar to that of the general population (although elsewhere it calls it very low because of possible germline mosaicism), and GeneReviews as presumably below 1% but higher than that of the population; and the PNDS reports transmission by affected people, which GeneReviews does not know of [3, 7]. The GeneReviews estimate, which takes germline mosaicism into account, is the more prudent for counselling.

8.3. Practical counselling points

In WAGR, the French PNDS recommends that the diagnosis be communicated at a dedicated consultation, by the clinician or geneticist who established it, with explanation of the diagnosis, the care and follow-up plan and genetic counselling, possibly with other team members (paediatrics, ophthalmology, paediatric neurology, oncology, nephrology, psychology, social services); the family should receive the contact details of the patient association, and a letter summarises for the corresponding physicians what was discussed [7]. The clinical genetics appointment should be as early as possible, to determine the mechanism and the risk of recurrence [7].

9. Special Considerations: Population Diversity and Epidemiology

Estimates of the frequency of aniridia vary with method and period. GeneReviews gives a prevalence of 1:64,000 to 1:96,000, based on Orphanet, which classifies isolated aniridia in the range of 1–9 per 100,000 [1, 4]. Two Scandinavian studies found 1 in 40,000 to 1 in 72,000 live births, in line with older epidemiological data from the United States; in Miller's 1964 series, the incidence considered was 1 in 50,000 [2]. The French national protocol gives 1/40,000 to 1/96,000 births, the European guidelines an overall incidence of 1:48,000 to 1:96,000, and the Japanese guideline 1/64,000 to 1/96,000, with no difference between sexes [9, 10, 20]. A 2023 review estimated more than 7,000 patients with congenital aniridia in the European Union [24]. The French WAGR PNDS cites an annual incidence of aniridia between 1/40,000 and 1/100,000 births, possibly underestimated and with no difference by ethnicity or sex, and a prevalence of 1:64,000 to 1:96,000 in a 2018 review; it also recalls that Wilms tumour accounts for 6–8% of paediatric cancers and that aniridia occurs in approximately 1 in 70 children with Wilms tumour; it does not present its own French epidemiological data [7]. Kit's British cohort showed no predilection by ethnicity or sex, but ethnicity was not documented in 41.9% of patients [26].

WAGR is estimated at 1:500,000 to 1:1,000,000 births [3, 4]. Among children with Wilms tumour, WAGR represented 0.75% (64 of 8,533) in the North American NWTSG between 1969 and 2002, and aniridia was recorded in 0.6% (47 of 7,842) of children in the SIOP 93-01 and 2001 protocols, although 12 of those records were classification errors [13, 35]. The proportion of WAGR among patients with aniridia depends on recruitment: chromosomal rearrangements disrupting PAX6 explain approximately one-third of sporadic aniridia and approximately 10% of all aniridia [2]; in the Homburg Registry, 20 of the 174 patients with a PAX6 alteration (10.7%) had WAGR(O) [32]; the Japanese guideline reports, from its systematic review, that nearly one-third of aniridia cases are part of WAGR [10]. Smith estimates that 1 in 70 patients with aniridia has Wilms tumour and that the tumour develops in one-third of patients with sporadic aniridia, but these numbers reflect series predating systematic molecular investigation [18]. These proportions describe hospital or registry series and should not be interpreted as population prevalences.

No founder effect for PAX6 or WAGR was identified in the consulted sources; the most frequent nonsense variants correspond to recurrent CpG dinucleotide mutations, not shared ancestral haplotypes, and GeneGuidance curation for WT1 records no founder variants or preferential geographical distribution [2, 5]. Genetic and clinical data derive mainly from cohorts in the United Kingdom, Scandinavia, Germany, France, Russia, Korea, India, the United States and Japan [2, 10, 24, 26, 30, 32]. The WAGR registry included participants from 16 countries or regions, mainly the United States (52) and United Kingdom (10), predominantly white (79 of 91), with 10 of the 48 who reported ethnicity identified as Hispanic or Latin American [8].

In Brazil, Cronemberger and colleagues, in Belo Horizonte, described monozygotic twin girls referred at 2 months for bilateral corneal opacity, with Peters anomaly and aniridia in both eyes and bilateral Wilms tumour in both, in whom exome sequencing identified a heterozygous 6,648,473 bp deletion at 11p13 [61]. GeneReviews also cites a Brazilian series of WT1-related disorders including a child with WAGR, a 46,XY karyotype and absent testes [3]. No Portuguese epidemiological data were found in the consulted sources; a clinical review by authors from Porto presents no national data [62].

For pregnancy in women with aniridia, the consulted sources contain no specific obstetric recommendations beyond preconception counselling and prenatal and preimplantation diagnostic options [1]. In children, choice of ocular hypotensive agents and off-label drug use require particular attention (section 7) [9, 49].

10. Illustrative Clinical Case

10.1. Family with a PAX6 missense variant, congenital cataract, glaucoma and pancreatic and olfactory manifestations

Buehne and colleagues, at Duke University (United States), described a proband, her mother and a brother with bilateral congenital cataracts removed at 2 months, 9 months and 3 months, respectively; the proband received an intraocular lens during surgery, the brother at age 7, and the mother has no implants [22]. Glaucoma was diagnosed in the proband at 6 months, the mother at 18–19 years and the brother at 6 years [22]. One or more of these relatives had anosmia, learning difficulties, pancreatitis in adulthood, pancreatic underdevelopment and infantile hypoglycaemia, and type 2 diabetes; the proband and mother reported reduced smell [22]. Examination of the proband under anaesthesia showed iris stromal hypoplasia and slight underdevelopment of the pupillary sphincter, without complete aniridia, with a Soemmering ring behind the intraocular lens (RetCam photograph in figure 2 of the article) [22].

A congenital cataract panel in 2016 was not diagnostic; in 2020, an 86-gene panel with copy-number variant calling identified the heterozygous variant NM_000280.3:c.178T>A p.(Tyr60Asn) (published numbering; = NM_001368894.2:c.220T>A p.(Tyr74Asn) in the MANE transcript) in exon 6 of PAX6, initially classified as of uncertain significance and later reclassified as likely pathogenic with the criteria PM1, PM2_P, PP3, PP4 and PP1; the variant was present in the affected mother and brother and absent in an unaffected brother, and genome sequencing in 2021 identified no other cause (figure 1, pedigree and chromatogram) [22]. The variant lies in the α3 (recognition) helix of the paired domain, a region with clustering of pathogenic missense variants, and is absent from gnomAD; it was deposited in ClinVar (SCV001906465) [22]. In the MANE transcript, it corresponds to NM_001368894.2:c.220T>A p.(Tyr74Asn) [21].

The case illustrates several points of this article: paired-domain missense variants may produce phenotypes without evident aniridia, dominated by cataract and glaucoma; the systemic PAX6 phenotype includes pancreatic and olfactory abnormalities; outdated panels may miss the diagnosis and reanalysis or new testing is useful; and PAX6 numbering differs between transcripts [1, 21, 22]. Reclassification relied on several independent criteria rather than segregation alone; the authors note that glaucoma after bilateral congenital cataract surgery before age 2 occurs in 6.7% in a meta-analysis, making the variant a plausible explanation for the familial glaucoma predisposition without demonstrating it [22].

10.2. Infant with WAGR and developmental glaucoma

Akagi and colleagues, at Kyoto University Hospital (Japan), described a boy with bilateral aniridia and developmental glaucoma in the right eye diagnosed at 1 month, who was subsequently diagnosed with WAGR [63]. Before age 1 he had undergone three right-eye trabeculotomies; at 15 months he was using three antiglaucoma eye drops, with IOP of 11 mmHg in the right eye and 24 mmHg in the left, bilateral macular hypoplasia, cup-to-disc ratios of 0.9 on the right and 0.6 on the left, corneal diameters of 14.0 × 13.5 mm and 13.5 × 13.0 mm, and right corneal opacification [63]. After three failed right and two failed left trabeculotomies, Baerveldt devices were implanted in both eyes, with the tube ligated, the plate in the superior temporal quadrant and the tube positioned between the cornea and lens and covered by donor sclera (figure 1; IOP course in figure 2) [63]. IOP stabilised 2–3 months later and the cup-to-disc ratios decreased to 0.3 and 0.4 at 14 months; at 15 months, the right tube touched the cornea and was repositioned with scleral fixation using a 10-0 polypropylene suture, without serious complications, maintaining IOP control for more than 2 years [63].

The case documents the poor response of aniridia-associated childhood glaucoma to angle surgery and the role of drainage devices, consistent with GeneReviews and the European and Japanese guidelines, and highlights the need to monitor contact between tube and endothelium in a growing eye [1, 9, 10, 63]. The article does not describe the deletion or renal course and therefore cannot support inferences concerning Wilms tumour risk [63].

11. Prognosis and Long-Term Follow-up

11.1. Visual and general prognosis in PAX6 aniridia

Visual prognosis depends mainly on foveal hypoplasia and, throughout life, progression of keratopathy, cataract and glaucoma [1, 9]. The European guidelines consider visual prognosis generally guarded in adulthood, with most patients having severe visual impairment or legal blindness, whereas GeneReviews emphasises that most retain useful vision with appropriate follow-up [1, 9]. In Jacobson's North American series, half of patients were legally blind at the end of follow-up, and final acuity was associated with initial acuity, glaucoma and the presence of cataract or keratopathy at presentation [30]. In the British longitudinal cohort, eyes never operated on retained better acuity, and carriers of missense variants had the mildest phenotype [26]. The Japanese guideline considers glaucoma irreversible and therefore central to follow-up [10].

11.2. Prognosis in WAGR

Survival from Wilms tumour is similar to that of the paediatric population with Wilms tumour: 95% at 4 years in the NWTSG and, in the SIOP-RTSG, 5-year event-free survival of 84.3% (95% CI, 72.4–98.1) and overall survival of 91.2% (95% CI, 82.1–100), with no relapses and with deaths attributable to toxicity and other causes [13, 35]. In the long term, however, estimated survival at 27 years from diagnosis was 48% in patients with WAGR versus 86% in the others, with five late deaths from end-stage kidney disease [35]. A follow-up study cited by Duffy estimated the 20-year incidence of end-stage kidney disease at 36% after unilateral tumour and 90% after bilateral tumour, usually in adolescence or later [8]. The French PNDS contrasts the old data (34–40% end-stage kidney disease in tumour survivors and 48% ± 17% survival at 27 years) with the 91% overall survival of the recent SIOP-RTSG series, and considers chronic kidney disease, often end-stage in adolescence or young adulthood, the main determinant of long-term prognosis [7, 13]. Life expectancy has not been formally studied; the IWSA knows of adults aged 40 and 50 years, and the oldest person with a molecular diagnosis described was 48 years old [3]. In the registry, 97.6% of participants lived in a situation of dependence, but most of those over 18 had completed secondary education and some were employed [8].

11.3. Follow-up schedule

Transition to adult care should ensure continuity of surveillance of glaucoma and of the ocular surface in aniridia, and of renal and cardiometabolic function in WAGR, with a summary of lifetime risk for adult nephrology [1, 3]; the French PNDS requires the transition to be prepared at the end of adolescence, to avoid interruptions in follow-up, IOP monitoring to continue throughout life, and the general practitioner, the local ophthalmologist and the reference centre to share care [7].

12. Summary and Final Considerations

Congenital aniridia is an autosomal dominant panocular disease caused in approximately 80–90% of cases by PAX6 loss-of-function variants and in approximately 10–20% by 11p13 deletions including the gene or its regulatory elements, some located within ELP4 [1]. Foveal hypoplasia and nystagmus determine initial vision; limbal deficiency keratopathy, cataract and glaucoma determine visual loss throughout life; and the phenotype includes central auditory, olfactory, sleep, neurodevelopmental and glycaemic abnormalities [1, 9]. Genotype–phenotype correlations (milder paired-domain missense variants and regulatory deletions; more severe keratopathy with extensive deletions and WAGR) guide group prognosis but do not predict individual progression [1, 24].

The most important decision in an infant with aniridia without a family history is to exclude an 11p13 deletion including WT1: CMA should be the first test, and renal ultrasound every 3 months should start while the result is awaited [1, 8, 11]. Once WAGR is confirmed, Wilms tumour risk is approximately one-half, with early tumours, frequently bilateral and with favourable histology, and ultrasound surveillance continues until age 7–8 according to European and North American recommendations, with lifelong renal surveillance because of chronic kidney disease risk, aggravated by nephrectomy [8, 11, 12, 36]. The genetic strategy combines copy-number analysis and sequencing, and genome sequencing resolves a substantial proportion of undiagnosed cases, including non-coding and structural variants [2, 40]. Reports should specify the MANE transcript (NM_001368894.2) and, when the literature uses NM_000280, both numbering systems [21].

Treatment remains supportive. The 2026 European and Japanese guidelines converge on the priority of medical treatment of glaucoma, caution with cataract and corneal surgery and the importance of low-vision rehabilitation, and diverge on artificial-iris contact lenses and angle surgery [9, 10]. The only randomised pharmacological trial, with oral ataluren in nonsense variants, did not meet its primary endpoint, and the drug is not authorised for aniridia in any jurisdiction; base-editing, drug-repurposing and tissue-engineering approaches are at the preclinical stage or in early trials [14, 53, 54, 56, 58]. Follow-up is lifelong, multidisciplinary and shared with families and patient associations, within the framework of ERN-EYE and the national reference centres, such as OPHTARA, whose 2021 PNDS is the only national protocol dedicated to WAGR [3, 7, 9, 15].

13. References

1. Kit V, Oluonye N, Moosajee M. PAX6 Aniridia Syndrome. 2003 May 20 [updated 2026 Jun 2]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. PMID: 20301534. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1360/
2. Hall HN, Williamson KA, FitzPatrick DR. The genetic architecture of aniridia and Gillespie syndrome. Hum Genet. 2019;138(8-9):881-98. doi:10.1007/s00439-018-1934-8. PMCID: PMC6710220.
3. George AM, Katz Z, Hathaway ER, Adams SA, Berlant Z, Duki BT, et al. WAGR Spectrum Disorder. 2026 Mar 12. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. PMID: 41818601. Available at: https://www.ncbi.nlm.nih.gov/books/NBK621298/
4. Orphanet. Isolated aniridia (ORPHA:250923; entry updated 2012) and WAGR syndrome (ORPHA:893; updated 2023) [Internet]. Paris: INSERM; accessed 2026 Oct 6. Available at: https://www.orpha.net/en/disease/detail/250923 and https://www.orpha.net/en/disease/detail/893
5. GeneGuidance. Gene and condition curation: PAX6, WT1, ELP4 [database]. Version of 26 September 2026; accessed 2026 Oct 6.
6. Han JC, Liu QR, Jones M, Levinn RL, Menzie CM, Jefferson-George KS, et al. Brain-derived neurotrophic factor and obesity in the WAGR syndrome. N Engl J Med. 2008;359(9):918-27. doi:10.1056/NEJMoa0801119. PMCID: PMC2553704.
7. Haute Autorité de Santé; Centre de Référence des Maladies Rares en Ophtalmologie OPHTARA. Protocole National de Diagnostic et de Soins (PNDS): Syndrome WAGR [text dated 2021 Jul 5]. Saint-Denis La Plaine: HAS; 2021 Oct 14. Available at: https://www.has-sante.fr/jcms/p_3283422
8. Duffy KA, Trout KL, Gunckle JM, Krantz SM, Morris J, Kalish JM. Results from the WAGR Syndrome Patient Registry: characterization of WAGR spectrum and recommendations for care management. Front Pediatr. 2021;9:733018. doi:10.3389/fped.2021.733018. PMCID: PMC8712693.
9. Romano D, Romano V, Daruich A, Ferrari G, Robert M, Semeraro F, et al.; ANIRIDIA-NET Clinical Guidelines Working Group; Aniridia-Europa and Geniris Working Group. Congenital aniridia: European COST action ANIRIDIA-NET guidelines for diagnosis, management and care. Acta Ophthalmol. 2026;104(3):267-79. doi:10.1111/aos.17587. PMID: 40892309. PMCID: PMC13058686.
10. Clinical Practice Guideline Development Committee for Aniridia of the "Research group on establishing standardized diagnosis and treatment of intractable corneal diseases" (Research on rare and intractable diseases, Health, Labour and Welfare Sciences Research Grants). Clinical practice guideline for aniridia. Jpn J Ophthalmol. 2026;70(2):441-71. doi:10.1007/s10384-025-01296-y. PMID: 41903000. PMCID: PMC13091811.
11. Hol JA, Jewell R, Chowdhury T, Duncan C, Nakata K, Oue T, et al. Wilms tumour surveillance in at-risk children: literature review and recommendations from the SIOP-Europe Host Genome Working Group and SIOP Renal Tumour Study Group. Eur J Cancer. 2021;153:51-63. doi:10.1016/j.ejca.2021.05.014. PMID: 34134020.
12. Kalish JM, Doros L, Helman LJ, Hennekam RC, Kuiper RP, Maas SM, et al. Surveillance recommendations for children with overgrowth syndromes and predisposition to Wilms tumors and hepatoblastoma. Clin Cancer Res. 2017;23(13):e115-22. doi:10.1158/1078-0432.CCR-17-0710. PMCID: PMC5538793.
13. Hol JA, Jongmans MCJ, Sudour-Bonnange H, Ramírez-Villar GL, Chowdhury T, Rechnitzer C, et al. Clinical characteristics and outcomes of children with WAGR syndrome and Wilms tumor and/or nephroblastomatosis: the 30-year SIOP-RTSG experience. Cancer. 2021;127(4):628-38. doi:10.1002/cncr.33304. PMCID: PMC7894534.
14. ClinicalTrials.gov. Study of Ataluren in Participants With Nonsense Mutation Aniridia (STAR). NCT02647359 [Internet]. PTC Therapeutics; registry with results, last updated 2022 May 27; accessed 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT02647359
15. European Reference Network for Rare Eye Diseases (ERN-EYE). Homepage and consensus documents [Internet]. Accessed 2026 Oct 6. Available at: https://www.ern-eye.eu/guidelines/
16. Miller RW, Fraumeni JF Jr, Manning MD. Association of Wilms's tumor with aniridia, hemihypertrophy and other congenital malformations. N Engl J Med. 1964;270:922-7. doi:10.1056/NEJM196404302701802. PMID: 14114111.
17. Riccardi VM, Sujansky E, Smith AC, Francke U. Chromosomal imbalance in the aniridia-Wilms' tumor association: 11p interstitial deletion. Pediatrics. 1978;61(4):604-10. PMID: 208044.
18. Jones KL, Jones MC, del Campo M. Aniridia–Wilms tumor association (WAGR syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 48–9.
19. Ton CC, Hirvonen H, Miwa H, Weil MM, Monaghan P, Jordan T, et al. Positional cloning and characterization of a paired box- and homeobox-containing gene from the aniridia region. Cell. 1991;67(6):1059-74. doi:10.1016/0092-8674(91)90284-6. PMID: 1684738.
20. Bremond-Gignac D, Robert M, Daruich A, Borderie V, Chiambaretta F, Valleix S, et al.; Groupe Rédacteur Relecteur PNDS Aniridie. [National protocol for diagnosis and care of congenital aniridia: summary for the attending physician]. J Fr Ophtalmol. 2022;45(6):647-52. doi:10.1016/j.jfo.2022.01.005. PMID: 35667788.
21. VariantValidator [online tool; version 4.0.1, database vvdb_2025_3]. Queries NM_000280.3:c.178T>A and NC_000011.10:g.31801740A>T (MANE Select) and MANE transcripts of PAX6 and WT1, on 2026 Oct 6. Available at: https://rest.variantvalidator.org
22. Buehne KL, Hart S, Williams B, Cohen JL. Novel PAX6 variant in a family with ophthalmologic, pancreatic, and olfactory features. Cold Spring Harb Mol Case Stud. 2022;8(1):a006149. doi:10.1101/mcs.a006149. PMCID: PMC8744493.
23. Firth HV, Hurst JA. Autosomal dominant (AD) inheritance. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 8–9.
24. Daruich A, Duncan M, Robert MP, Lagali N, Semina EV, Aberdam D, et al. Congenital aniridia beyond black eyes: from phenotype and novel genetic mechanisms to innovative therapeutic approaches. Prog Retin Eye Res. 2023;95:101133. doi:10.1016/j.preteyeres.2022.101133. PMCID: PMC11062406.
25. Szentmáry N, Suiwal S, Amini M, Fries FN, Náray A, Latta L, et al. Aniridia-associated keratopathy: clinical and molecular mechanisms of disease progression and emerging therapeutic targets. Acta Ophthalmol. 2026;104(6):604-15. doi:10.1111/aos.70086. PMCID: PMC13460364.
26. Kit V, Cunha DL, Hagag AM, Moosajee M. Longitudinal genotype-phenotype analysis in 86 patients with PAX6-related aniridia. JCI Insight. 2021;6(14):e148406. doi:10.1172/jci.insight.148406. PMCID: PMC8410060.
27. Firth HV, Hurst JA. Anterior segment eye malformations. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 62–4.
28. Firth HV, Hurst JA. Optic nerve hypoplasia. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 270–1.
29. Firth HV, Hurst JA. Cataract. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 88–91.
30. Jacobson A, Mian SI, Bohnsack BL. Clinical outcomes and visual prognostic factors in congenital aniridia. BMC Ophthalmol. 2022;22(1):235. doi:10.1186/s12886-022-02460-5. PMCID: PMC9131660.
31. Firth HV, Hurst JA. Glaucoma. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 428–9.
32. Obst J, Fries FN, Amini M, Náray A, Munteanu C, Stachon T, et al. Systemic diseases in patients with congenital aniridia: a report from the Homburg Registry for Congenital Aniridia. Ophthalmol Ther. 2025;14(2):433-45. doi:10.1007/s40123-024-01084-w. PMCID: PMC11754556.
33. Turner JT, Brzezinski J, Dome JS. Wilms Tumor Predisposition. 2003 Dec 19 [updated 2022 Mar 24]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. PMID: 20301471. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1294/
34. Firth HV, Hurst JA. Wilms tumour. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 620–2.
35. Breslow NE, Norris R, Norkool PA, Kang T, Beckwith JB, Perlman EJ, et al.; National Wilms Tumor Study Group. Characteristics and outcomes of children with the Wilms tumor-aniridia syndrome: a report from the National Wilms Tumor Study Group. J Clin Oncol. 2003;21(24):4579-85. doi:10.1200/JCO.2003.06.096. PMID: 14673045.
36. Tracy ET, Leraas H, Olson L, Shamberger RC, Ehrlich PF. Wilms tumor characteristics, surgical management, outcomes, and chronic kidney disease in children with WAGR syndrome: a report from the International WAGR Syndrome Association survey. Pediatr Blood Cancer. 2024;71(9):e31172. doi:10.1002/pbc.31172. PMID: 38956810.
37. Firth HV, Hurst JA. Nystagmus. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 252–3.
38. Firth HV, Hurst JA. Hypospadias. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 186–7.
39. Firth HV, Hurst JA. Cerebellar anomalies. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 92–5.
40. Hall HN, Parry D, Halachev M, Williamson KA, Donnelly K, Campos Parada J, et al. Short-read whole genome sequencing identifies causative variants in most individuals with previously unexplained aniridia. J Med Genet. 2024;61(3):250-61. doi:10.1136/jmg-2023-109181. PMCID: PMC7615962.
41. AURAGEN consortium. Unveiling ocular developmental disorders through short-read whole-genome sequencing. Eur J Hum Genet. 2026;34(10):1573-84. doi:10.1038/s41431-026-02160-4. PMID: 42448966. PMCID: PMC13634046.
42. Jones KL, Jones MC, del Campo M. Peters-plus syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 838–9.
43. Jones KL, Jones MC, del Campo M. Encephalocraniocutaneous lipomatosis. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 740–1.
44. Jones KL, Jones MC, del Campo M. Goltz syndrome (focal dermal hypoplasia). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 754–7.
45. Jones KL, Jones MC, del Campo M. Microphthalmia–linear skin defects syndrome (MIDAS syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 758–9.
46. Oteuliyeva M, Kyrykbayev D, Zhurgumbayeva G, Smailova D, Ordabayev I, Brusati L. Enhancing medical care for aniridia through a public health lens: a scoping review of organizational models, patient management, and quality of life assessment. Front Public Health. 2026;14:1890217. doi:10.3389/fpubh.2026.1890217. PMCID: PMC13543156.
47. European Medicines Agency. Ikervis (ciclosporin): EPAR — medicine overview and product information [Internet]. Amsterdam: EMA; accessed 2026 Oct 6. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/ikervis
48. Tóth G, Liu S, Li S, Váncsa S, Paczkó M, Kói T, et al. Outcomes of penetrating keratoplasty and Boston type 1 keratoprosthesis in aniridia-associated keratopathy: a systematic review and meta-analysis. Am J Ophthalmol. 2026;282:405-15. doi:10.1016/j.ajo.2025.11.027. PMID: 41276206.
49. U.S. Food and Drug Administration. ALPHAGAN P (brimonidine tartrate ophthalmic solution) 0.1% and 0.15% — prescribing information (NDA021262/NDA021770; version of 2025 Sep 18) [Internet]. openFDA; accessed 2026 Oct 6. Available at: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22ALPHAGAN+P%22
50. Tóth G, Liu S, Li S, Váncsa S, Paczkó M, Kói T, et al. Outcomes of trabeculectomy and glaucoma drainage device surgery in congenital aniridia-associated glaucoma: a systematic review and meta-analysis. Ophthalmol Ther. 2025;14(10):2323-43. doi:10.1007/s40123-025-01215-x. PMID: 40736869. PMCID: PMC12413375.
51. U.S. Food and Drug Administration. Premarket approval P170039: CustomFlex Artificial Iris (HumanOptics AG), decision of 2018 May 30 [Internet]. openFDA, PMA database; accessed 2026 Oct 6. Available at: https://api.fda.gov/device/pma.json?search=pma_number:P170039
52. U.S. Food and Drug Administration. Drugs@FDA (openFDA): search for ataluren, no records [Internet]. Accessed 2026 Oct 6. Available at: https://api.fda.gov/drug/drugsfda.json
53. European Medicines Agency. Translarna (ataluren): EPAR — authorisation history, renewal opinions and non-renewal decision [Internet]. Amsterdam: EMA; accessed 2026 Oct 6. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/translarna
54. PTC Therapeutics, Inc. Form 10-Q, quarterly report for the period ended June 30, 2026 [Internet]. Washington (DC): U.S. Securities and Exchange Commission; 2026 Jul 30. Available at: https://www.sec.gov/Archives/edgar/data/1070081/000107008126000017/tmb-20260630x10q.htm
55. ClinicalTrials.gov. A Phase 2 Open Label Extension Study in Participants With Nonsense Mutation Aniridia. NCT04117880 [Internet]. PTC Therapeutics; status: withdrawn; accessed 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT04117880
56. ClinicalTrials.gov. RAFT - Clinical Trial of RAFT for Aniridia Related Keratopathy. NCT05044598 [Internet]. University College London; status: completed (2025 Apr 17); accessed 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT05044598
57. Djayet C, Bremond-Gignac D, Touchard J, Secretan PH, Vidal F, Robert MP, et al. Formulation and stability of ataluren eye drop oily solution for aniridia. Pharmaceutics. 2020;13(1):7. doi:10.3390/pharmaceutics13010007. PMCID: PMC7822034.
58. Mirjalili Mohanna SZ, Korecki AJ, Kazemian P, Lengyell TC, Adair BA, Leavitt BR, et al. A novel minimally humanized mouse model of aniridia for preclinical evaluation of CRISPR gene editing strategies. Front Genome Ed. 2026;8:1910209. doi:10.3389/fgeed.2026.1910209. PMCID: PMC13631079.
59. Koc AC, Kocak G, Kahveci B, Mardinoglu A, Karakülah G, Utine CA, et al. Patient-derived cornea organoids as drug repurposing models for aniridia-associated keratopathy. Life Sci. 2026;402:124586. doi:10.1016/j.lfs.2026.124586. PMID: 42447954.
60. Lima Cunha D, Sarkar H, Eintracht J, Harding P, Zhou JH, Moosajee M. Restoration of functional PAX6 in aniridia patient iPSC-derived ocular tissue models using repurposed nonsense suppression drugs. Mol Ther Nucleic Acids. 2023;33:240-53. doi:10.1016/j.omtn.2023.06.016. PMCID: PMC10362734.
61. Cronemberger S, Albuquerque ALB, Silva ACSE, Zanini JLSS, da Silva AHG, Barbosa LF, et al. Bilateral Peters' anomaly, aniridia and Wilms tumour (WAGR syndrome) in monozygotic twins. Acta Paediatr. 2024;113(6):1420-5. doi:10.1111/apa.17160. PMID: 38363039.
62. Calvão-Pires P, Santos-Silva R, Falcão-Reis F, Rocha-Sousa A. Congenital aniridia: clinic, genetics, therapeutics, and prognosis. Int Sch Res Notices. 2014;2014:305350. doi:10.1155/2014/305350. PMCID: PMC4897488.
63. Akagi T, Yoshikawa M, Nakanishi H, Yoshimura N. A case of WAGR syndrome in association with developmental glaucoma requiring bilateral Baerveldt glaucoma implants and subsequent tube repositioning. Clin Ophthalmol. 2015;9:1081-4. doi:10.2147/OPTH.S80444. PMCID: PMC4474385.