Waardenburg Syndromes

Auditory-pigmentary neurocristopathies (types I to IV, PCWH and Klein-Waardenburg) due to a deficit of melanocytes in the skin, hair, iris and stria vascularis of the cochlea, with variable involvement of the enteric nervous system, myelin, limbs and gonadotropic axis; genes PAX3, MITF, SOX10, EDNRB, EDN3, SNAI2 and KITLG.

Literature and regulatory information update: 10 October 2026.

0. Initial Reference Table

The table brings together the entities grouped in this article. The abbreviation WS (Waardenburg syndrome) and the numbering WS1 to WS4 follow international nomenclature; PCWH designates peripheral demyelinating neuropathy, central dysmyelinating leukodystrophy, Waardenburg syndrome and Hirschsprung disease (peripheral demyelinating neuropathy, central dysmyelinating leukodystrophy, Waardenburg syndrome and Hirschsprung disease). The MANE Select transcript (Matched Annotation from NCBI and EMBL-EBI) is the reference transcript agreed between NCBI and EMBL-EBI; OMIM (Online Mendelian Inheritance in Man) numbers identify phenotypes and genes. Orphanet prevalences are classes or mean values with the validation status indicated by the database itself [1–7].

1. Introduction

1.1. The original description and clinical delineation

In 1951, Petrus J. Waardenburg published in the American Journal of Human Genetics the description of “a new syndrome combining developmental anomalies of the eyelids, eyebrows and nose root with pigmentary defects of the iris and head hair and with congenital deafness” [16]. According to the Smith synopsis, Waardenburg found the pattern in 1.4% of children with congenital deafness and estimated an incidence of about 1 in 42,000 in the Netherlands [17]. The feature the author placed in the foreground, lateral displacement of the inner canthi and lacrimal puncta (dystopia canthorum), would come to separate type 1 from the others. In 1978, Arias and Mota studied two large pedigrees and showed that variable expressivity of dystopia could be confused with non-penetrance; they introduced the biometric W index, which distinguishes individuals with and without dystopia and reliably identifies dystopia in more than 96% of carriers [18].

Clinical heterogeneity was established in the following decades. In 1981, Shah and colleagues, in India, described the association of white forelock, iris pigmentary change and long-segment Hirschsprung disease, which they proposed as a possible variant of Waardenburg syndrome; hence the eponym Waardenburg-Shah, now WS4 [19]. Klein reported in 1950 the association of limb anomalies with the auditory-pigmentary signs, the origin of type 3 (Klein-Waardenburg); Goodman and colleagues documented in 1982 bilateral musculoskeletal hypoplasia, flexion contractures, fusion of the carpal bones and syndactyly in two Yemenite Jewish siblings, a family whose follow-up by Sheffer and Zlotogora confirmed dominant inheritance [20, 21]. In 1992, the Waardenburg Consortium published its first report: with linkage data from 41 families with WS1 and 3 with WS2 and six chromosome 2q markers, it placed a locus between ALPP and FN1 and found that this locus explained about 45% of the families in the sample; in the course of the work the major and minor diagnostic criteria still in use were defined [22]. Liu, Newton and Read, in Manchester, personally studied 81 individuals from 21 families with WS2 and proposed their own criteria for that type [23].

1.2. Milestones of molecular discovery

In 1992, two groups simultaneously published the first pathogenic variants in Nature: Tassabehji and colleagues, in Manchester, mapped a WS gene to distal 2q and, through homology with the Splotch mouse (Pax-3), demonstrated mutations in the human homologue PAX3 in families with WS [24]; Baldwin and colleagues, in Boston, identified an exonic mutation in the HuP2 gene (PAX3) in a large Brazilian family, present in all affected members and absent in 50 controls [25]. That family was part of the Recife pedigrees studied by da-Silva, with 73 affected individuals in two families, in whom telecanthus was the only constant anomaly [26]. In 1993, Hoth and colleagues found two variants in the paired domain of PAX3, one in a WS1 family and the other in a WS3 family, showing that the two types are allelic [27].

WS2 was linked to 3p12.3–p14.1 and, in 1994, Tassabehji, Newton and Read identified splice site mutations of MITF, the homologue of the mouse microphthalmia gene, in two families [28]. The endothelin pathway entered the story through Hirschsprung disease: Puffenberger and colleagues identified in 1994, in a Mennonite kindred, the EDNRB variant p.Trp276Cys, with a dosage effect [29]; in 1996, Edery and colleagues and Hofstra and colleagues described homozygous EDN3 variants in patients with Waardenburg-Shah [30, 31]. In 1998, Pingault and colleagues found SOX10 variants in four families with WS4, after the murine gene Sox10 had been identified as the cause of the Dominant megacolon phenotype [32]. In 1999, Inoue and colleagues described the first patient with Pelizaeus-Merzbacher-type leukodystrophy, Charcot-Marie-Tooth type 1 neuropathy and Waardenburg-Hirschsprung due to a SOX10 variant [33], and in 2004 the same group explained the difference between PCWH and WS4 by the fate of the mutant mRNA in the face of nonsense-mediated decay (NMD) [34].

The following years broadened the spectrum. Sánchez-Martín and colleagues described homozygous SNAI2 deletions in two patients with WS2 (2002) and heterozygous deletions of the same gene in piebaldism (2003) [35, 36]; OMIM later reclassified the WS2D deletion as a variant of uncertain significance [10]. Bondurand and colleagues demonstrated, in 2007, heterozygous deletions of the SOX10 locus in WS4 and WS2 [37]. Pingault and colleagues showed in 2013 that SOX10 loss-of-function variants explain about one third of cases of Kallmann syndrome with hearing loss [38]. Zazo Seco and colleagues linked KITLG to unilateral or asymmetric hearing loss and to a family with WS2 in 2015 [39], and Issa and colleagues established in 2017 that heterozygous EDNRB variants cause WS2 with incomplete penetrance [40].

1.3. Nomenclature and reclassifications

The classification into four types is clinical. WS1 is defined by dystopia canthorum; WS2, by its absence; WS3, by musculoskeletal anomalies of the limbs; and WS4, by Hirschsprung disease [1, 9]. The correspondence with genes is partial: the review by Pingault and colleagues (2010) brought together six genes (PAX3, MITF, EDN3, EDNRB, SOX10, SNAI2) with different frequencies in each type [41], and the systematic review by Song and colleagues concluded that the gene predicts the auditory phenotype better than the clinical type [13]. There are cases with a WS1 phenotype and a variant in MITF or biallelic EDNRB [42, 43], and a Japanese study of 16 families found variants in MITF, SOX10 or EDNRB, and not in PAX3, in 61% of individuals classified as WS1 by the W index [44]. In 2022, Salah and colleagues proposed reserving the designation “Waardenburg” for patients with a monoallelic PAX3 variant (with or without limb anomalies) and grouping the biallelic ones under “Klein syndrome”; this is a nosological proposal without consensus [8]. ABCD syndrome (albinism, black lock, cell migration disorder of the neurocytes of the gut and deafness), caused by homozygous EDNRB p.Arg201Ter, was reinterpreted by its own authors as an expression of Shah-Waardenburg syndrome, a reading adopted by OMIM for WS4A, which records AD and AR inheritance [45, 46]. The designation “Waardenburg anophthalmia” corresponds to ophthalmo-acromelic syndrome due to biallelic SMOC1, an entity unrelated to the auditory-pigmentary syndromes [47].

1.4. Definition of the entities within the scope

Orphanet defines Waardenburg syndrome as a rare genetic multiple congenital anomaly characterised by hearing loss and defects of structures derived from the neural crest, including pigmentary anomalies of the eyes, hair and skin, with four clinical phenotypes [1]. WS1 combines congenital hearing loss, minor defects of neural crest structures and dystopia canthorum [2]. WS2 is the subtype with variable degrees of hearing loss and pigmentary anomalies, without dystopia canthorum [3]. WS3 is a rare subtype with limb anomalies, congenital hearing loss and pigmentary changes [4]. WS4 combines the auditory-pigmentary picture with Hirschsprung disease (aganglionic megacolon) [5]. PCWH adds to WS4 neurological manifestations of variable severity [6]. Tietz syndrome is characterised by congenital profound bilateral sensorineural hearing loss and generalised “albinoid” hypopigmentation of the skin, eyes and hair [7, 48]. Read and Newton summarised the common basis: the auditory-pigmentary syndromes result from the physical absence of functional melanocytes in the skin, hair, eyes or stria vascularis of the cochlea [49].

2. Molecular Genetics and Pathophysiology

2.1. Neural crest, melanocytes and stria vascularis

Melanocytes derive from the neural crest, produce eumelanin and pheomelanin in melanosomes and transfer them to keratinocytes; depigmentation (amelanosis) reflects the absence or lack of function of melanocytes, and hypopigmentation their partial reduction [50]. In the cochlea, the intermediate cells of the stria vascularis derive from melanocytes and are necessary for inner ear homeostasis; the lack of these cells compromises the stria vascularis and produces sensorineural hearing loss [9, 11]. The otopathology of a patient with WS1 and a PAX3 variant, described by Merchant and colleagues, documented defective melanocyte migration or function and defective development of the stria vascularis [11, 51]. The same logic explains the other manifestations: absence of melanocytes in the skin produces leucoderma; in the follicle, the white forelock and premature greying; in the iris stroma and choroid, the hypoplastic blue iris, heterochromia and hypopigmented ocular fundus [49, 52].

Waardenburg syndromes are therefore neurocristopathies [11]. The melanocyte differentiation pathway has a known hierarchy: PAX3 and SOX10 synergistically activate the MITF promoter, by direct binding to a proximal region with sites for both factors, and the mutant proteins of either lose this capacity [53]. MITF and SOX10 then activate DCT and TYR, genes of melanin synthesis [9]. In the Dom mouse, Sox10 dysfunction compromises Mitf expression and melanocyte development and survival, which explains the auditory-pigmentary signs common to the three genes [53]. In the gut, colonisation by vagal neural crest cells occurs between 5 and 12 weeks of gestation, in a craniocaudal direction [54]; SOX10 maintains enteric progenitors and regulates RET, EDNRB and its own expression, and EDN3/EDNRB signalling prevents premature differentiation of melanoblast and enteric neuroblast precursors [9, 55]. Loss of this signalling leaves the skin, hair, iris and stria vascularis without melanocytes, and the distal gut without ganglia [55]. SGRT presents the syndrome as a consequence of neural tube closure defects in the third week; this formulation does not correspond to the mechanism described in the molecular sources, which is a defect of specification, migration, proliferation or survival of neural crest cells [11, 56].

2.2. PAX3

PAX3 (2q36.1; OMIM 606597; HGNC:8617) is one of the nine human PAX genes and encodes a transcription factor with a paired domain and homeodomain, expressed in the early embryo and an essential regulator of muscle and neural crest derivatives, including melanocytes [11, 55]. It regulates proliferation, migration and apoptosis and sits at the top of the melanocyte pathway, where it transactivates MITF in synergy with SOX10; its role in myogenesis probably explains the muscular hypoplasia and contractures of WS3 [55]. The Oxford Desk Reference still places the gene at 2q35, a cytoband superseded in current annotation [11, 57]. The MANE Select transcript validated in VariantValidator is NM_181458.4 (PAX3d isoform); GeneReviews numbers variants on NM_181457.4 and part of the literature uses NM_000438 [11, 58].

The mechanism of WS1 is loss of function through haploinsufficiency (ClinGen haploinsufficiency score of 3) [11, 55]. Functional studies add detail: WS1 variants alter binding and regulation of reporters with MITF or TYRP1 promoters, revealing variant-dependent modes of DNA recognition [59]; altered subnuclear localisation and mobility are indicators of dysfunction [60]; homeodomain substitutions may destabilise folding or interaction with DNA [61]; and, in zebrafish, a possible dominant negative effect linked to loading of PAX3 on mitotic chromosomes, mediated by arginine methylation, has been described [62]. Partial or whole PAX3 deletions do not produce a more severe phenotype than intragenic variants [11, 63].

Tassabehji and colleagues screened 134 families or individuals with auditory-pigmentary syndromes and found PAX3 variants in 20 of 25 families with definite WS1 and in 1 of 2 with WS3, and in none of the 23 families with definite WS2 or of 36 other neurocristopathies; variant types included missense at conserved amino acids of the paired domain or homeodomain, splicing variants, nonsense variants and frameshift indels [64]. In the HGMD (Human Gene Mutation Database), cited by Zhan and colleagues, there were 226 pathogenic PAX3 variants, with a predominance of missense and nonsense (about 51%) and concentration in exons 2 to 6, which encode the paired domain and homeodomain [65].

Genotype-phenotype correlation is limited. DeStefano and colleagues observed pigmentary changes more frequently with variants deleting the homeodomain than with missense variants or deletions including the paired domain, without correlation for hearing loss [66]. There are two codon correlations: p.Asn47His in WS3 [27] and p.Asn47Lys in craniofacial-deafness-hand syndrome (CDHS), which combines a flat facial profile, widely spaced eyes, a hypoplastic nose with slit-like nares, absent or small nasal bones, a small maxilla, ulnar deviation of the hands and sensorineural hearing loss [11, 67]. The gene-disease validity of CDHS has divergent classifications in ClinGen/GenCC submissions (from limited to definitive), and the GeneReviews author acknowledges genetic heterogeneity [11, 55]. Homozygosity changes the phenotype: in a consanguineous Turkish family, both parents heterozygous for p.Tyr90His had WS1 and the homozygous 18-month-old son had typical WS3, with dystopia canthorum, partial albinism and upper limb defects; it was the third report of PAX3 homozygosity with WS3 [68]. WS3 is not, however, always recessive: Sheffer and Zlotogora described two siblings with Klein-Waardenburg whose father and paternal aunt were also affected, documenting dominant transmission [21], and the Oxford Desk Reference records at least two parent-to-child transmissions [57]. The somatic PAX3-FOXO1 fusion (gain of function) of alveolar rhabdomyosarcoma is a tumour event; patients with this tumour do not have WS1 and WS1 is not associated with oncological risk [11].

2.3. MITF

MITF (3p13; OMIM 156845) encodes a bHLH-LZ (basic helix-loop-helix leucine zipper) transcription factor, the master regulator of melanocyte survival and differentiation and of melanosome biogenesis; it binds M-box and E-box motifs of genes such as TYR and TYRP1 and participates in the differentiation of mast cells, osteoclasts and retinal pigment epithelium [55]. mi mice have hypopigmentation and, depending on the allele, microphthalmia, deafness, osteopetrosis and mast cell defects [28]. The gene has several isoforms: the MANE Select is NM_001354604.2 (MITF-A), but much of the literature numbers variants on the melanocytic isoform MITF-M (NM_000248), and the c. and p. numbering differs between them [58].

Grill and colleagues characterised 24 variants: most of those associated with WS2A and Tietz syndrome neither bind DNA nor activate melanocytic promoters, which constitutes loss of function (ClinGen haploinsufficiency score of 3); p.Arg203Lys and p.Ser298Pro had normal activity and are probably neutral; melanoma variants bind DNA normally and act through another mechanism [55, 69]. Tietz syndrome, with congenital profound deafness and generalised (not patchy) hypopigmentation, AD inheritance and complete penetrance, results from missense variants in the basic region of MITF; in the original family on reassessment, it contrasts with the patchy depigmentation and variable deafness of WS2 [48]. Orphanet records two families [7]. GeneGuidance curation places Tietz syndrome at the severe end of the MITF axis [55].

Biallelic variants cause COMMAD syndrome (coloboma, osteopetrosis, microphthalmia, macrocephaly, albinism and deafness), described by George and colleagues in children of parents with monoallelic MITF variants; it was the first human MITF phenotype with microphthalmia [70]. In October 2026, Pampanos and colleagues described a Greek infant with a COMMAD-type phenotype, who died in the early neonatal period, with ectopic ossification, cardiac fibrosis and fibroelastosis, interstitial lung disease, ectopic renal tissue, hepatic steatosis and haemosiderosis, carrying two truncating MITF variants (p.Arg321Ter, previously described, and p.Lys379Ter, novel) with phase undetermined because parental testing was declined; the transcript is not stated in the abstract and the numbering has not been standardised [71]. When both parents have mild WS2 due to MITF, homozygosity may produce a WS4 picture [17].

Morell and colleagues described an apparent digenic form of WS2 with ocular albinism: a 1 bp deletion in exon 8 of MITF and the TYR allele p.Arg402Gln, a functional polymorphism with moderately reduced tyrosinase activity, in homozygosity or heterozygosity in all individuals with ocular albinism [72].

The MITF variant p.Glu318Lys (E318K), NM_000248.4:c.952G>A p.(Glu318Lys) on the MITF-M isoform, corresponds to NM_001354604.2:c.1273G>A p.(Glu425Lys) on the MANE transcript (conversion in VariantValidator) [58]. It lies in a consensus SUMOylation site; severely reduced SUMOylation increases binding to the HIF1A promoter, the number of occupied loci and cell clonogenicity, migration and invasion, i.e. a gain of function [55, 69]. Yokoyama and colleagues obtained a lod score of 2.7 in 31 melanoma families under a dominant model and association in Australian and British case-control samples, with over-representation in cases with a family history, multiple primary melanomas or both, more naevi and non-blue eyes [73]. Bertolotto and colleagues described a more than 5-fold risk of melanoma, renal cell carcinoma or both in genetically enriched families [74]. This variant does not cause Waardenburg syndrome and the gene-disease curations of WS2A and Tietz syndrome document no oncological risk [55].

2.4. SOX10

SOX10 (22q13.1; OMIM 602229; MANE NM_006941.4) encodes a 466-amino-acid transcription factor of the SOX family (HMG box, high-mobility group), of the SOXE group with SOX8 and SOX9; it has a 79-amino-acid HMG domain that binds the minor groove of DNA, with two nuclear localisation signals and one export signal, a dimerisation domain, a central transactivation domain and a C-terminal one [9]. SGRT places the gene at 20q13.1, which is an error; the cytoband is 22q13.1 [5, 56]. Two exon numbering systems coexist: the historical one, with five exons (two non-coding, ATG in exon 3 and stop codon in exon 5), and that of the reference transcript NM_006941, with four exons; the difference has consequences for reading reports [9].

SOX10 is expressed in neural crest derivatives (melanocytes, enteric nervous system, Schwann cells, olfactory ensheathing cells) and in tissues not derived from it (inner ear, oligodendrocytes) [9]. In Schwann cells it participates in myelinating differentiation (via OCT6/POU3F1 and EGR2); in oligodendrocytes it activates myelination with MYRF, OLIG1, TCF4 and CHD7; in olfactory ensheathing cells, its loss impairs the migration of neurons producing GnRH (gonadotropin-releasing hormone), which explains Kallmann syndrome [9, 38].

There are two main mechanisms. Haploinsufficiency results from deletions and from truncating variants degraded by NMD and produces WS2 or WS4; the dominant negative effect results from truncated proteins that escape NMD and produces PCWH [34, 55]. PCWH correlates with variants in the last coding exon and in the last 45 nucleotides of the penultimate one; the earlier the truncation in the last exon, the more severe the picture [9]. Non-stop variants, with an 82-amino-acid extension that includes a deleterious domain, produce PCW or PCWH through toxic gain of function, whereas the extension in another frame described as p.(Tyr460LeufsTer42) does not [9]. Missense variants at Gln174 and Pro175 form nuclear foci with p54NRB and sequester the normal protein, and are associated with a progressive central and peripheral neurological phenotype [9]. 22q13.1 duplications including SOX10 may cause PCWH and, in some cases, sex reversal; a triplication has been described in a 46/47,XX girl without sex reversal [9]. Heterozygous deletions of the locus have been described in WS4 and WS2, and some patients with WS2 and a deletion had a PCWH-type neurological phenotype [37].

In a series of about 300 independent cases, just over half of the variants were truncating (nonsense or frameshift), about one third non-truncating (missense or small in-frame indels) and about 10% complete or partial copy-number variations (CNVs); splicing, start codon and non-stop variants were rare [9]. Truncating variants occur at any point except the extreme C-terminus; missense variants cluster in the HMG domain. A rare or novel missense variant in the HMG domain has a high probability of pathogenicity; outside the HMG domain, even if rare and predicted pathogenic, it should be treated cautiously as a variant of uncertain significance (VUS), with a possible second hotspot in the dimerisation domain (valines 76, 79 and 80) [9]. Truncating variants predominate in WS4 and PCWH; the proportion of missense variants increases in WS2 and especially in Kallmann syndrome, where there is a tendency to overrate rare variants owing to incomplete penetrance and digenism [9]. According to GeneGuidance, ClinVar listed 175 pathogenic, 100 likely pathogenic and 168 VUS variants, figures that reflect testing volume and not frequency [55].

2.5. EDNRB and EDN3

EDNRB (13q22.3; OMIM 131244; MANE NM_001122659.3) encodes the endothelin receptor type B, a G protein-coupled receptor that binds endothelins 1, 2 and 3 without selectivity and signals through the phosphatidylinositol-calcium pathway [55]. EDN3 (20q13.32; OMIM 131242; MANE NM_207034.3) encodes a secreted peptide, cleaved from a precursor, which is the receptor ligand; the variant described by Hofstra and colleagues probably impairs proteolytic processing of preproendothelin [31, 55]. The Edn3/Ednrb mouse models have megacolon and white spotting [55].

Inheritance is semidominant, with a dosage effect. Biallelic variants produce complete WS4; heterozygotes may have no signs or isolated features, especially isolated Hirschsprung disease, with reduced penetrance [12, 55]. Svensson and colleagues described a heterozygous frameshift EDN3 variant in sporadic short-segment Hirschsprung disease without WS, and Pingault and colleagues a heterozygote with complete Waardenburg-Hirschsprung, showing that the correlation is not always linear [12, 55]. The Mennonite variant EDNRB NM_001122659.3:c.828G>T p.(Trp276Cys), in the fifth transmembrane helix, retains partial function (partial reduction of ligand-induced calcium transients) and conferred a Hirschsprung disease risk of 74% in homozygotes and 21% in heterozygotes; the kindred behaved as a multigenic trait [29]. ClinVar classifies it with conflicting interpretations (VCV 16633) [75]. Issa and colleagues, starting from an exome in a WS2 index case, found 6 heterozygous EDNRB variants in a WS2 cohort, with dominant inheritance of incomplete penetrance and defective localisation or signalling in functional studies; they estimated that EDNRB explains 5–6% of WS2 and that, previously, MITF and SOX10 explained only about 30% [40]. In the review by Pingault and colleagues (2010), about 70% of families with homozygous or compound heterozygous EDNRB segregated as a fully recessive disease; in the others, some heterozygous relatives had isolated Hirschsprung disease, constipation or depigmentation, and three families with dominant WS4 of incomplete penetrance had been described [41].

2.6. SNAI2 and KITLG

SNAI2 (SLUG; 8q11.21; OMIM 602150; MANE NM_003068.5) encodes a zinc-finger transcriptional repressor of the Snail family, which binds E-boxes, represses CDH1, BRCA2 and ITGA3 and participates in neural crest generation and migration as an effector of the epithelial-mesenchymal transition, downstream of TWIST1 and TGFB1 [55]. Human evidence is limited to whole-gene deletions: homozygous in two unrelated patients with WS2, with absence of the product and with MITF transactivating the SLUG promoter [35], and heterozygous in 3 of 17 patients with piebaldism without a KIT variant [36]. The Slug mouse has anaemia, infertility, a white forehead blaze and ventral depigmentation [36]. Pingault and colleagues searched for SNAI2 point variants and deletions in the French WS2 cohort and found none, concluding that the gene plays at most a minor role [41]. OMIM no longer considers the WS2D form established and has reclassified the SNAI2 deletion as a variant of uncertain significance [10]. In 2025, Xue and colleagues described in a Chinese family the SNAI2 variant c.230C>G in six carriers from three generations (zygosity not stated in the abstract), with moderate to severe intellectual disability, severe anxiety, mild depression and social dysfunction, without hearing loss or heterochromia, and Snai2-/- mice with cognitive deficit, depigmented coat and arrested development of the astrocyte lineage; the phenotype lacks the cardinal signs of WS and the association is not validated [76].

KITLG (12q21.32; OMIM 184745; MANE NM_000899.5) encodes the ligand of the KIT receptor. It is associated with three distinct diseases [55]. DFNA69 hearing loss (OMIM 616697) is a congenital, stable, non-syndromic, unilateral or asymmetric sensorineural hearing loss, with AD inheritance of reduced penetrance, defined by the truncating variant NM_000899.5:c.286_303delinsT p.(Ser96Ter), without detectable soluble KITLG, and by p.(His67_Cys68delinsArg) [39, 58]. The French national series of Serey-Gaut and colleagues (14 symptomatic individuals from 8 families) showed unilateral, bilateral symmetrical or asymmetric hearing loss, predominantly at low frequencies when not profound, incomplete penetrance and non-specific pigmentary changes in only 3 of the 14; KITLG was the predominant gene in isolated unilateral hearing loss of early, non-sudden onset [55]. WS2F (OMIM 619947, classified as AR in OMIM) is based on the heterozygous variant NM_000899.5:c.310C>G p.(Leu104Val), which segregated with WS2 in a small family and whose transmembrane isoform reaches the membrane, with reduction of the soluble isoform, interpreted as a dominant negative or gain-of-function mechanism [39, 55]; a biallelic series described hypomelanosis with sensorineural hearing loss, symmetrical “stocking and glove” distribution, progressive repigmentation or generalised hypomelanosis, and proposed that biallelic loss of function produces generalised hypomelanosis and variants with residual function a WS2-type or piebaldism picture [55]. Familial progressive hyper- and hypopigmentation (OMIM 145250), AD with reduced penetrance, has hyperpigmented and café-au-lait macules, hypopigmented macules and lentigines [55].

2.7. Gene-disease validity

The ClinGen hearing loss expert panel (Hearing Loss Gene Curation Expert Panel) classified the validity of the associations between 2017 and 2024 [77].

The practical consequence is that a heterozygous SNAI2 finding in a panel report should be interpreted with caution in the light of the KIT, PAX3, MITF and SOX10 results [55], and that the heterozygous associations of KITLG and EDNRB/EDN3 have limited evidence [77].

2.8. Genetic architecture by clinical type

The proportions for each gene vary between sources, because they depend on the inclusion criterion, period and detection methods.

The role of SOX10 in isolated Hirschsprung disease is minimal: in 229 cases, a single frameshift variant was found, inherited from an asymptomatic mother [9]. There is no evidence that RET causes WS4 [12].

2.9. Variants, nomenclature and standardisation

The variants in the table were standardised in VariantValidator (GRCh38) by the data collection agents and in this write-up; conversions between isoforms were not made by hand [58].

The MITF variants p.(Leu341ArgfsTer18) of the twins with a WS1 phenotype [42] and p.Arg321Ter and p.Lys379Ter of the COMMAD case [71] were published without a declared transcript; given the isoform dependence, they remain in the source notation. The EDNRB variant p.Arg201Ter of ABCD syndrome is likewise in the original notation [45].

2.10. Penetrance, expressivity, digenism and modifiers

In WS1, penetrance is probably almost complete, but expressivity is highly variable, even within the same family [11]. SGRT reports a “penetrance of 36–58%” in type 1, a value corresponding to the frequency of isolated features and not to the penetrance of the genotype [11, 56]. Read and Newton emphasise that severity cannot be predicted, even with an identified variant [49]. For SOX10, hearing loss is the almost fully penetrant sign, pigmentary defects are not always present and confirmed incomplete penetrance is very rare, and should prompt targeted clinical reassessment [9]. Even so, in a Chinese family with SOX10 p.(Arg43Ter), the carrier father was phenotypically normal, the aunt had hearing loss without heterochromia, the sister had hearing loss with heterochromia (WS2 phenotype) and the proband WS4C [79]. For EDN3 and EDNRB, penetrance in heterozygotes is reduced and dose-dependent [29, 55]. For KITLG, penetrance is incomplete for both hearing loss and hyper/hypopigmentation, so a carrier relative without signs is expected [55].

Digenism and dosage modify the phenotype. Yang and colleagues described a family in which the parent with WS1 had a PAX3 variant and the parent with WS2 an MITF variant; the child with both variants had far more pigmentary changes (white forelock, white eyebrows and eyelashes, leucoderma) [11, 84]. Tassabehji and colleagues observed that gene dosage has a greater effect in humans than in mice [64]. TYR p.Arg402Gln acts as a modifier in WS2 with ocular albinism [72].

3. Clinical Spectrum and Manifestations by System

3.1. Variability and data sources

All forms show marked variability, from minimal pigmentary changes to profound hearing loss, even within the same family [49, 57]. Feature frequencies come from series with different biases. Liu and colleagues compared 81 individuals from 21 families with WS2 studied personally with 60 patients from 8 families with WS1 and 253 published cases [23]. Pardono and colleagues, in São Paulo, studied 59 patients from 37 families (30 with WS1, 21 with WS2 and 8 isolated cases without telecanthus), with 23 craniofacial measurements and 300 stratified controls, and estimated the frequencies of the eight cardinal signs in 461 affected individuals with WS1 and 121 with WS2, combining their own and literature data [85]. Tamayo and colleagues, in a screening programme in Colombia, found similar percentages, but all 95 affected individuals had hearing loss and came from the institutionalised deaf population, which introduces ascertainment bias [11, 86]. Newton reviewed the clinical features of the different Waardenburg syndromes [87]. The systematic review by Song and colleagues brought together 73 articles and 417 patients with a genotype, and is subject to publication bias towards genotyped cases [13].

Source: GeneReviews, Table 4, based on Liu 1995, Pardono 2003 and Tamayo 2008 [11, 23, 85, 86].

3.2. Hearing and the vestibular system

Hearing loss is sensorineural, congenital, typically non-progressive, unilateral or bilateral, mild to profound, and most often profound and bilateral (>100 dB); laterality varies between and within families [11, 17]. In the review by Song and colleagues, hearing loss was present in 71.0% of the 417 genotyped patients, predominantly bilateral and sensorineural, with a significant difference between types and between genes [13].

The divergence between 71% and “about 50%” results from the denominator: the systematic review includes genotyped patients, many ascertained through hearing loss, whereas family series include relatives identified by pigmentary or facial signs. In MITF carriers, the Oxford Desk Reference estimates the risk of clinically significant hearing loss at about one third; the others have minor pigmentary anomalies and some have mild hearing loss without clinical significance [57]. Progression is also debated: the review by Pingault and colleagues (2010), citing Hildesheimer and colleagues (1989), states that in WS2 hearing loss is progressive in 70% of cases and that there is no typical audiogram shape, whereas the Oxford Desk Reference and GeneReviews describe the hearing loss as usually non-progressive; the same review places the prevalence of temporal bone anomalies described in the series between 0 and 50%, with a predominance of enlarged vestibular aqueduct and semicircular canal malformations, and associates delayed walking with bilateral vestibular areflexia with canal malformation [11, 41, 57].

In WS1, Madden and colleagues described temporal bone anomalies in patients with hearing loss: enlargement of the vestibular aqueduct and the upper vestibule, narrowing of the porus of the internal auditory canal and modiolar hypoplasia [11, 90]. The Smith text mentions aplasia of the posterior semicircular canal in WS1 [17]. Vestibular symptoms (vertigo, dizziness, imbalance) may exist even without hearing loss [11, 91], and delayed motor milestones are common in neurologically normal deaf children owing to vestibular involvement [57].

In SOX10 variants, inner ear malformations are the rule. In the series of Elmaleh-Bergès and colleagues, with 15 patients, temporal bone anomalies were bilateral in all: mainly agenesis or hypoplasia of at least one semicircular canal with an enlarged vestibule and a small cochlea, sometimes deformed, with normal partition; three patients had no cochlear nerve, bilaterally in two [55, 92]. On magnetic resonance imaging (MRI), there was agenesis of the olfactory bulbs in 7 of 8, hypoplastic or absent lacrimal glands in 11 of 14, hypoplastic parotids in 12 of 14 and white matter signal change in 7 of 13 [55, 92]. Pingault and colleagues consider these malformations apparently exclusive to patients with SOX10, highly penetrant when read by a specialist radiologist, although some patients do not have them; complete vestibular function has not yet been studied [9]. In a Korean cohort of 610 patients (825 ears) with cochlear implants, cochlear duct length was reduced in cochlear nerve deficiency, SOX10-associated WS, enlarged vestibular aqueduct and incomplete partition type III, with the greatest reduction in SOX10 WS [93]. SOX10 hearing loss is typically prelingual, non-progressive, profound and bilateral, but may be moderate, asymmetric or unilateral, and some patients appear to have isolated hearing loss until a targeted history reveals minor signs [9].

3.3. Pigmentation of the hair, skin and eyes

The white forelock is the most common hair anomaly in WS1, present in about 45% of cases; it may be present at birth or appear later, typically in adolescence, and may repigment; it is usually located in the midline, but may appear elsewhere, and red and black forelocks have been described [11]. Most patients with WS1 have a white forelock or greying of the scalp before 30 years of age [11, 22]. The eyebrows and eyelashes may be hypopigmented [11, 17]. Considering WS1 and WS2 together, Pingault and colleagues estimate the white forelock in at least one third of patients, greying before 30 years in one third of those without a forelock, the hypoplastic brilliant-blue iris in 10%, heterochromia in about 30% and cutaneous pigmentary changes in 10–20%, mainly on the extremities of the limbs, abdomen and chest; synophrys is much more frequent in WS1 (85%) than in WS2 (25%), and in heterochromia the ocular fundus is also heterogeneous, blond or albinoid in the regions corresponding to the hypoplastic iris, without visual impairment [41].

Congenital leucoderma appears on the face, trunk or limbs, frequently with hyperpigmented borders, and may accompany an adjacent forelock [11]. The Smith text describes the set as partial albinism: white forelock, isochromic pale blue irides with hypoplastic stroma or heterochromia, areas of vitiligo, white patches of hair outside the forelock and mottled peripheral retinal pigmentation [17].

Ocular findings include complete or segmental heterochromia (two colours in the same iris, typically brown and blue), hypoplastic blue or brilliant-blue iris and hypopigmentation of the iris and choroid, with a more sectoral than diffuse pattern [11, 52]. Visual acuity is the same as in the general population [11]. In 2026 two clinical images were published: one of a 69-year-old woman with a clinical diagnosis of WS2, bilateral hypopigmentation and hyperpigmentation of the iris and choroid, congenital deafness, heterochromia and a grey forelock since birth [94], and one of foveal hypoplasia in WS, known only by its title [95].

In Tietz syndrome, hypopigmentation is generalised and “albinoid”, without a patchy pattern [7, 48]. In biallelic KITLG WS2F, hypomelanosis may have a symmetrical “stocking and glove” distribution, repigment progressively or be generalised, and pigmented macules have been described [55].

3.4. Face and skull

In WS1, the Smith text describes lateral displacement of the inner canthi with short palpebral fissures and lateral dystopia of the lacrimal puncta, a broad high nasal root with hypoplastic alae nasi, thick eyebrows with medial flare that may meet (synophrys) and a broad mandible; occasionally, patent metopic suture, strabismus, a rounded nasal tip, full lips with an accentuated “cupid's bow”, a smooth philtrum, cleft lip and palate and anisocoria [17]. The aberrant pattern of the mid-eyebrow reflects altered growth of the underlying facial structures between 10 and 16 weeks of fetal life [96]. In the Recife families studied by da-Silva (73 affected individuals), telecanthus was the only constant anomaly; a prominent nasal root appeared in 78%, thick eyebrows with synophrys in 76%, sensorineural hearing loss in 67%, hypopigmented skin lesions in 55%, premature greying in 44% and a white forelock in 29%, with a rounded or square nasal tip, hypoplastic alae and a smooth philtrum [26, 97]. Cleft lip and palate have been described in several families, with no determinable percentage [11]. The Smith figures illustrate the phenotype: Figure 1 (p. 360) shows lateral displacement of the medial canthi, a broad nasal root with hypoplastic alae, medial flare of the eyebrows and hypochromic irides; Figure 2 (p. 361) shows a 6-year-old girl with WS1 due to a PAX3 variant, with telecanthus and short palpebral fissures, anisochromia of the irides, thick eyebrows with medial flare, a broad high nasal root, rounded nasal tip, smooth philtrum, white forelock and patchy hypopigmentation of the leg [17]. SGRT presents a woman with a frontal white forelock, heterochromia, profound bilateral deafness and dystopia canthorum and her son with the same picture (Fortaleza, Brazil), and patients with deafness and heterochromia without dystopia canthorum, with WS2 [56].

3.5. Nervous system

Neural tube defects occur in about 1–2% of carriers of PAX3 variants, mainly heterozygous and occasionally homozygous [17]. Spina bifida has been described in several families, including a prenatal diagnosis in a family with WS1 [11, 98]; Lemay and colleagues found a de novo nonsense PAX3 variant in a patient with myelomeningocele and WS1 [99]; Chen and colleagues described a partial PAX3 deletion (exons 1 to 4, including the promoter), of paternal origin, in a previous pregnancy with spina bifida and in a second pregnancy whose newborn had a normal phenotype at birth [100]. SGRT includes Waardenburg syndrome among the genetic syndromes with neural tube closure defects [101].

The SOX10 forms involve the peripheral and central nervous system. PCWH has symptoms that depend on the severity of the myelin defects: motor and cognitive delay, a cerebral palsy-like picture, ataxia, spasticity, congenital nystagmus, hyporeflexia, distal sensory deficits and distal muscle atrophy [9]. Some forms of WS2E have cognitive impairment, myelination defects and ataxia [55], and in SOX10 WS4 neurological symptoms with demyelination may be progressive [54]. OMIM also records reduced tear production, hypotonia, nystagmus and signs of dysautonomia in one of the European patients of Touraine and colleagues, and an 11-year-old Spanish child with PCWH, cochlear hypoplasia and progressive sensorineural hearing loss, which contradicts the rule of stable hearing loss [102]. Pingault and colleagues described a patient with hypomyelinating neuropathy, hearing loss and chronic intestinal pseudo-obstruction, without Hirschsprung disease or pigmentary changes, due to a de novo SOX10 deletion [102]. Bogdanova-Mihaylova and colleagues described a 32-year-old Irish man with a novel heterozygous SOX10 missense variant, congenital deafness, Hirschsprung disease, heterochromia, foot deformity and length-dependent sensorimotor neuropathy with intermediate conduction velocities [103].

3.6. Olfactory and gonadotropic axis

Agenesis of the olfactory bulbs is frequent in SOX10 WS [9, 92]. Most patients with SOX10 Kallmann syndrome have hearing loss, and Kallmann syndrome and WS are not mutually exclusive: patients with an initial diagnosis of WS have been diagnosed with hypogonadism at puberty, and anosmia and hypogonadism are probably underdiagnosed, as they are difficult to recognise before puberty [9]. Yang and colleagues described two cases with a SOX10 variant, early auditory intervention (cochlear implant at 3 years in one case, hearing aids from 2 years in the other) and age-appropriate language, both with delayed puberty and biochemical hypogonadotropic hypogonadism [104].

3.7. Limbs and skeleton

WS3 combines the features of WS1 with muscular hypoplasia of the upper limbs, flexion contractures, fusion of carpal bones and syndactyly, with occasional camptodactyly and hearing loss in 57% [11, 17]. The Oxford Desk Reference describes it as an extreme presentation of type 1, with flexion contractures of the fingers [57]. The Turkish child homozygous for p.Tyr90His had, at 18 months, dystopia canthorum, partial albinism and upper limb defects [68], and the siblings described by Sheffer and Zlotogora had blepharophimosis and bilateral flexion contractures of the fingers [21]. In WS1, Sprengel anomaly, supernumerary vertebrae and ribs and scoliosis have occasionally been described [17].

3.8. Digestive system

Hirschsprung disease defines WS4 [5, 17]. In SOX10 forms, its frequency approaches 100%; in EDNRB and EDN3 forms it is common [12]. Some patients with WS4 or PCWH have hypoganglionosis or chronic intestinal pseudo-obstruction (CIPO), probably neurogenic, instead of Hirschsprung disease [9]. Ersson and colleagues compared two Swedish newborns with four controls: in PCWH due to SOX10 p.(Lys257Ter) there was a reduction of glial cells (SOX10) and hypoganglionosis (HuC/D) in the small and large intestine; in WS4 with EDN3 p.(Arg158Cys), severe reduction of glia and aganglionosis with a skip segment; interstitial cells of Cajal (CD117) were preserved in both [81]. In MITF carriers, the risk of Hirschsprung disease is small or nil [57]. In WS1, imperforate anus has occasionally been described [17]. Hirschsprung disease has its own neonatal presentation, risk of enterocolitis and surgical treatment.

3.9. Other organs

In WS1, ventricular septal defect, multicystic dysplastic kidney and absence of the vagina and uterine adnexa have occasionally been described [17]. Pingault and colleagues recorded two cases of biallelic SOX10 deletion, probably two pregnancies of the same couple with both parents heterozygous with classic WS, with a severe polymalformative fetal phenotype; the maternal deletion included 18 further genes [9]. Solia-Nasser and colleagues described the dental phenotype in an extended family with WS1, according to SGRT [56].

3.10. Natural history and treated disease

The auditory-pigmentary picture is congenital and, in WS1, hearing loss does not progress [11]. Morbidity and mortality are concentrated in WS4, owing to the risk of death from enterocolitis [56], and in the neurological SOX10 forms, whose course may be progressive [9, 54]. With early detection of the hearing deficit and appropriate rehabilitation, neuropsychological development is normal [56]; cochlear implant results are presented in section 7. In the African systematic review by Aboagye and colleagues (84 cases), mean age at report was 17.5 years [105].

4. Clinical Recognition and Warning Signs

4.1. In the newborn and infant

Congenital hearing loss is detectable by newborn hearing screening [11]. In 2019 the Joint Committee on Infant Hearing (JCIH) set the 1-3-6 goals (screening completed by 1 month, audiological diagnosis by 3 months and early intervention by 6 months) and recommends that programmes already meeting them aim for a 1-2-3 timetable [106]. The same document lists among the risk factors for late-onset or progressive hearing loss a family history of hearing loss with onset in childhood (factor 1), craniofacial conditions (factor 8) and the more than 400 syndromes associated with atypical hearing thresholds (factor 9), and expressly mentions Waardenburg syndrome among the frequent syndromes with sensorineural hearing loss [106]. In Table 1 of the document, for infants who pass screening, diagnostic audiological assessment is recommended by 9 months for factor 1 (with subsequent monitoring according to the aetiology of the familial hearing loss and caregiver concern), for factor 8, which expressly includes the white forelock among conditions at birth, and for factor 9 (subsequent monitoring according to the natural history of the syndrome or concerns); parental concern always requires additional assessment [106]. In a newborn with identified WS, the search for hearing loss should be immediate, because identification and rehabilitation before the critical period for language determine speech and language [55].

The emergency sign is intestinal. A newborn with SOX10-linked WS or suspected WS4 who does not pass meconium in the first 48 hours, has abdominal distension, bilious vomiting or signs of enterocolitis needs urgent paediatric surgical assessment, with contrast enema and suction rectal biopsy; Hirschsprung-associated enterocolitis may be fatal and diagnostic delay is avoidable harm [54, 55]. The guideline of ERNICA (European network for rare inherited and congenital digestive disorders) indicates rectal biopsy in the presence of the triad of delayed passage of meconium beyond 24 hours in the term newborn, abdominal distension and bilious vomiting, and establishes that the threshold for biopsy is lowered by the presence of a syndrome associated with Hirschsprung disease or a family history (level of evidence III, strong recommendation, 100% agreement) [15]. About 90% of patients with Hirschsprung disease present in the neonatal period, with distension in more than 90%, vomiting in more than 85% and failure to pass meconium in the first 24 hours in more than 60% [15].

4.2. In the child with hearing loss

In the deaf child, the Oxford Desk Reference recommends a three-generation family history with questions directed at pigmentary anomalies (Waardenburg, piebaldism) and an examination looking for dystopia canthorum, heterochromia or hypoplastic irides, white forelock and subtle cutaneous pigmentary changes [57]. The 2022 American College of Medical Genetics and Genomics (ACMG) guideline points out that hearing loss occurs in only some patients with WS and that the diagnosis may be missed if pigmentary or gastrointestinal changes are not specifically asked about; it recommends that all children and adolescents with hearing loss of unknown aetiology be assessed for syndromes by a clinical geneticist [88]. When the hair shows no forelock, it is worth asking whether it has been dyed; the forelock may appear in adolescence or repigment [11]. Isolated unilateral hearing loss of early, non-sudden onset, with or without non-specific pigmentary changes, should suggest KITLG [55].

4.3. In the presence of pigmentary lesions or Hirschsprung disease

In the presence of hypo- or depigmented lesions, it is important to ask whether they have been present since birth, whether they progress, whether there is hearing loss, neurological signs, Hirschsprung disease or chronic constipation since birth, and to look for a white forelock, premature greying and heterochromia, with Wood's lamp and clinical photography [50]. In any patient with Hirschsprung disease, pigmentary anomalies (white forelock, heterochromia, cutaneous streaks), a family history of hearing loss and pigmentary anomalies and neurological signs should be sought [54].

4.4. Signs pointing to SOX10 and neurological forms

The combination of anosmia, absent or arrested puberty and hearing loss points to SOX10, even without pigmentary signs [38, 55]. Hypotonia, congenital nystagmus, motor delay, ataxia, spasticity or peripheral neuropathy in an infant with WS4 suggest PCWH [9]. On imaging, bilateral agenesis or hypoplasia of the semicircular canals with an enlarged vestibule and small cochlea, in the appropriate clinical context, strongly suggests SOX10 WS [55, 92].

4.5. Examination of the face

The telecanthus of WS1 (inner canthi and lacrimal puncta displaced laterally) is distinguished from true hypertelorism (increased interpupillary distance) and from epicanthic folds by measurement of the inner intercanthal, outer intercanthal and interpupillary distances [107]. The W index is calculated with a ruler in the clinic [55]. The sources consulted do not describe specific biochemical or genetic newborn screening for Waardenburg syndromes; the observational Early Check study, of expanded newborn screening in the USA, is active but not recruiting [108].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Waardenburg Consortium clinical criteria

The clinical diagnosis of WS1 requires two major criteria or one major and two minor [11, 22]. SGRT reproduces the same rule [56]. According to Liu and colleagues, the diagnosis of WS2 is secure in the presence of a family history of congenital hearing loss and pigmentary changes, provided that measured ocular distances exclude dystopia canthorum; sensorineural hearing loss (77%) and heterochromia (47%) are the two most important indicators of WS2 [23]. Types 3 and 4 are defined by the criteria of WS1 or WS2 plus, respectively, limb anomalies and Hirschsprung disease [9].

5.2. The W index and its limitations

The W index is calculated with distances in millimetres: a = inner intercanthal; b = interpupillary; c = outer intercanthal. X = (2a − (0.2119c + 3.909))/c; Y = (2a − (0.2479b + 3.909))/b; W = X + Y + a/b. Dystopia is present if W >1.95 [11]. If the mean family W index is below 1.95, the diagnosis is WS2 [11]. GeneReviews provides a spreadsheet [11]. The index originates from the work of Arias and Mota, who applied empirical probabilities of dystopia and blepharophimosis in the general population [18].

The limitations are well documented. In the Brazilian series of Pardono and colleagues, the discriminant function of the four ocular measurements (inner and outer intercanthal, interpupillary and lower lacrimal distance) classified patients perfectly, as did the family mean W index; the individual W index correctly classified 93% of WS2, but only 60% of WS1 [85]. In 16 Japanese families, 61% of individuals classified as WS1 by W >1.95 had variants in MITF, SOX10 or EDNRB, and not in PAX3, so the threshold is not valid in all ethnic groups [11, 44]. In 2026, Mahoney and colleagues retrospectively reviewed patients at the Children's Hospital of Philadelphia between 2000 and 2023: 56 patients with clinical or genetic WS and 1744 patients without WS with a measured W index [89]. With the 1.95 threshold, comparing 41 patients with WS and a recorded W index with the 1744 without WS, sensitivity was 63.4% (95% CI 48.7–78.2%), specificity 86.8% (95% CI 85.2–88.4%), positive predictive value 10.2% and negative predictive value 99.0%, with an odds ratio of 11.4 for WS given a raised index; ROC (receiver operating characteristic) analysis showed that a threshold of 2.05 would raise specificity to 95.1% and the positive likelihood ratio from 4.6 to 9.4, at the cost of a sensitivity of 46.3% [89]. Patients with PAX3 had a mean W index of 2.248, all above 1.95 except one (1.84), whereas the mean of patients with SOX10 and MITF was below 1.95, with one MITF patient at 2.18 [89]. The authors conclude that the index is more useful for ruling out WS than for establishing it and propose raising the threshold to 2.05, accepting more false negatives and the bias of a population referred for hearing loss; the article's own table gives 86.3% specificity for the 1.95 threshold, slightly different from the 86.8% in the text [89]. In practice, the clinical subtype does not identify the gene, and a multigene panel is the appropriate test when the picture is not typical [55].

5.3. Molecular tests

GeneReviews describes two approaches. When the phenotype is distinctive, PAX3 sequencing is performed (small intragenic, missense, nonsense and splicing variants) and, if negative, targeted deletion and duplication analysis (quantitative PCR, i.e. quantitative polymerase chain reaction; long-range PCR; MLPA, multiplex ligation-dependent probe amplification; or targeted microarray); alternatively, a multigene panel with PAX3 and the genes of the differential diagnosis, which limits VUS and findings unrelated to the phenotype, and whose genes and sensitivity vary between laboratories. When the phenotype cannot be distinguished from other hereditary hearing losses, exome (more widely used) or genome sequencing is performed [11].

These proportions refer to probands with a clinical diagnosis of WS1 and are not equivalent to the clinical sensitivity of the test in an unselected child with hearing loss. In the Philadelphia cohort, the genetic cause was identified in 38 of 56 patients with WS (67.9%): PAX3 in 18, SOX10 in 8, MITF in 7, EDNRB in 5 and MITF with EDNRB in one patient, with no variants in KITLG or EDN3 [89]. In other cohorts, overall yields were lower: in Brazil, Bocángel and colleagues studied 49 probands with a clinical diagnosis (24 WS1 and 25 WS2) by Sanger sequencing of PAX3, MITF, EDN3, EDNRB, SOX10 and SNAI2 and MLPA of PAX3, MITF and SOX10 in selected cases, and identified the cause in 19 (38.8%); MLPA identified 6 exonic deletions, i.e. the cause in 12.2% of probands and 31.6% of causal variants, most of them sporadic [82]. In Hong Kong, Poon and colleagues obtained 65% molecular diagnosis in 30 patients from 19 families, with single-gene, panel or research genome testing [109]. In a Chinese cohort of 90 probands with WS, cited by Zhan and colleagues, PAX3 explained 13 (14.4%), of which 3 were de novo [65].

For SOX10, the strategy must include copy-number analysis, because deletions of the locus cause WS2 and WS4; negative sequencing does not exclude the gene [37, 55]. The position of a truncating variant relative to NMD distinguishes restricted WS from PCWH [9, 55]. The report should state the transcript: for SOX10, the historical exon numbering differs from that of NM_006941 [9]; for MITF, numbering varies between the MANE isoform and the melanocytic isoform [58]. SOX10 variants increasingly appear in atypical contexts (isolated hearing loss, neurodevelopmental disorders) and are difficult to classify without a typical phenotype [9].

5.4. Limitations of testing and interpretation

A negative result does not exclude the diagnosis [88]. Sequencing alone does not detect exonic deletions, which account for about 6% of causes in WS1 and 31.6% of causal variants in the Brazilian series [63, 82], nor deletions of the SOX10 locus [37]; and leucocyte DNA may not detect somatic mosaicism and does not detect exclusively germline mosaicism [11]. The ACMG recommends knowing the coverage, analytical sensitivity and variant types detected by each panel [88]. Pathogenic and likely pathogenic variants are equivalent for clinical decision-making, according to the criteria of the ACMG and the Association for Molecular Pathology (AMP); a VUS in PAX3 neither confirms nor excludes the diagnosis [11, 110]. In the presence of a VUS, segregation studies in relatives may help to reclassify it [88]. The report should state the gene, the transcript with version, HGVS (Human Genome Variation Society) c. and p. nomenclature, zygosity, ACMG/AMP class and parental origin, when studied. The case in section 10 illustrates the reasoning: PAX3 c.788dup received the criteria PVS1 (null variant in a gene where loss of function is the mechanism) and PM2_supporting (absence from population databases), with a classification of likely pathogenic [65].

In most patients with WS1, the diagnosis is established by physical examination, and molecular testing confirms it when the clinical picture is inconclusive [11]. WS1 is one of the few conditions in which non-molecular assessment is diagnostic and not merely supportive; testing identifies the gene and is necessary because the W index performs poorly in some populations [55]. SGRT considers molecular testing the definitive diagnosis [56].

5.5. Non-molecular investigations

Audiological assessment includes newborn hearing screening, auditory brainstem responses (ABR) and pure-tone audiometry [55]. The Oxford Desk Reference grades hearing loss as mild (20–40 dB), moderate (40–60 dB), severe (60–80 dB) and profound (>80 dB), placing conversational speech between 45 and 60 dB [57]. Computed tomography (CT) and MRI of the temporal bone are indicated when malformation is suspected or for cochlear implant candidacy; CT better assesses bony structures, and MRI shows the inner ear, cochlear nerves and brain without radiation [88, 106]. When SOX10 is suspected, the same MRI may show the olfactory bulbs, lacrimal and parotid glands, white matter and cochlear nerve [9, 55]. Biopsy of depigmented skin shows absence of melanocytes, which confirms the nature of the lesion but does not identify the gene [55]. In WS3, orthopaedic assessment and hand radiographs document carpal fusion, syndactyly and contractures [55]. In PCWH, nerve conduction studies and brain MRI characterise the neuropathy and dysmyelination; in the Kallmann presentation, smell testing, basal gonadotropins and sex steroids [55].

When Hirschsprung disease is suspected, the diagnosis is histological: suction or open rectal biopsy, equally accurate if there is sufficient submucosa, taken from the posterior or lateral wall at least 2 cm from the pectinate line or 3 cm from the anal verge, with 2 to 3 fragments in the suction technique, and confirmed before pull-through surgery (level III, strong recommendation, 100% agreement) [15]. Histology assesses the absence of ganglion cells with calretinin and/or peripherin and acetylcholinesterase activity; contrast enema and anorectal manometry are less accurate than an adequate biopsy [15, 55]. In the European Paediatric Surgeons' Association (EUPSA) survey, all respondents performed rectal biopsy (61% suction, 39% open full-thickness), with haematoxylin-eosin in 77%, acetylcholinesterase in 74% and calretinin in 31%, contrast enema in 96% and manometry in 31% [111].

5.6. Investigation of the deaf child according to the 2022 ACMG guideline

The ACMG guideline structures the assessment in five recommendations [88]. The first applies to all newborns and infants with confirmed hearing loss: prenatal history (cytomegalovirus, rubella, syphilis, thalidomide, retinoids), neonatal history (prematurity, low birth weight, hypoxia, hyperbilirubinaemia, sepsis, ototoxic drugs) and postnatal history (meningitis, trauma, noise, ototoxic drugs); audiometric characterisation (type, age at onset, progression or fluctuation, laterality, symmetry, severity, configuration, vestibular dysfunction, auditory neuropathy); a three-generation pedigree with first- and second-degree relatives with hearing loss or associated signs, such as pigmentary anomalies; and dysmorphological examination with attention to pigmentation. The second establishes that, in the presence of syndromic signs, pre-test counselling and, with consent, confirmatory testing (single gene, hearing loss panel, exome, genome, karyotype, microarray or CNV analysis by next-generation sequencing) be performed, with study of other organs and a screening plan. The third, without syndromic signs, proposes a comprehensive hearing loss panel with the genes recommended by ClinGen and, if negative, exome or genome; temporal bone imaging is complementary, especially with an uncertain diagnosis, implant candidacy, auditory neuropathy or progressive hearing loss, weighing radiation and sedation; cytomegalovirus is sought as early as possible and within the first 3 weeks of life, by PCR on saliva (preferred) or urine, and thereafter on the newborn screening dried blood spot. The fourth recommends referral to a multidisciplinary centre and, without an aetiological diagnosis, genetic reassessment every 3 years. The fifth requires that positive, negative or inconclusive results be communicated with genetic counselling [88]. Hearing loss panels registered in the GTR (Genetic Testing Registry) have on average 121 genes (23–308), with a yield of about 40% for comprehensive panels and 30–35% for the exome; in unilateral hearing loss, the yield is 1–5% [88]. The electrocardiogram, thyroid function, urinalysis and renal ultrasound may await the genetic result if there is no clinical concern [88]. The Oxford Desk Reference adds parental audiograms, ophthalmological assessment (refraction and fundus) in all deaf children and an electrocardiogram for the QT interval in severe or profound hearing loss [57].

5.7. Practical algorithm

5.8. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing are possible once the familial variant has been identified, but they do not predict the manifestations or severity [11]. Reproductive details are in section 8.

6. Differential Diagnosis

6.1. Comparative table

6.2. Distinguishing the types from each other

The distinction between WS1 and WS2 depends on dystopia canthorum, measured and not estimated; the family mean W index is more reliable than the individual value [11, 85]. The distinction is of limited value for choosing the test: in the Japanese families, most individuals with W >1.95 had variants in other genes [44], and the gene predicts hearing loss better than the type [13]. In WS4, the origin of the Hirschsprung disease points to the gene: heterozygous SOX10 with almost 100% Hirschsprung disease, frequently de novo, or EDN3/EDNRB, usually biallelic [12]. Bilateral semicircular canal malformations, agenesis of the olfactory bulbs or neuropathy point to SOX10 [9].

6.3. Piebaldism and Tietz syndrome

Piebaldism shares the white forelock and absence of pigmentation of the medial forehead and eyebrows; depigmentation of the chest, abdomen and limbs is common, hyperpigmented borders are characteristic and heterochromia and hearing loss are rare [11]. Most cases are due to KIT; heterozygous SNAI2 deletions have been described in 3 of 17 patients without a KIT variant [36, 55]. Tietz syndrome is distinguished by generalised hypopigmentation, without patches, and by profound hearing loss with complete penetrance [48]. SGRT asks whether Tietz syndrome is a severe WS2; GeneGuidance curation places it at the severe end of the MITF axis [55, 56].

6.4. Albinism and cellular trafficking disorders

Oculocutaneous albinism has diffuse hypopigmentation, photophobia, nystagmus and visual impairment, without hearing loss [56]. Vesicle trafficking disorders (Griscelli, Hermansky-Pudlak, Chediak-Higashi) and Menkes and Vici diseases combine hypopigmentation with immunodeficiency, bleeding diathesis or neurological disease, which sets them apart from the auditory-pigmentary picture [113]. Hermansky-Pudlak syndrome likewise enters the differential diagnosis.

6.5. Hearing loss with endocrine or neurological signs

Kallmann syndrome with hearing loss may be the only expression of a SOX10 variant; without pigmentary changes, the combination of Kallmann syndrome, hearing loss and semicircular canal anomalies requires consideration of mild forms of CHARGE syndrome [9]. PCWH is distinguished from other leukodystrophies and hereditary neuropathies by its association with the auditory-pigmentary and intestinal picture [9, 33].

6.6. Isolated hearing loss

Isolated unilateral hearing loss of early onset, with incomplete penetrance in the family, suggests KITLG [55]. Profound bilateral hearing loss with an enlarged vestibular aqueduct suggests Pendred syndrome, and branchial pits or fistulae with renal anomalies suggest branchio-oto-renal syndrome [57]. The ACMG emphasises that WS may be confused with non-syndromic hearing loss if pigmentation is not actively sought [88].

7. Treatments and Risk Management Strategies

7.1. Principles and initial assessment

There is no disease-modifying therapy for any of the Waardenburg syndromes [11, 55, 114]. Treatment is directed at the manifestations and multidisciplinary, and hearing loss is the main disabling symptom [56]. After a diagnosis of WS1, GeneReviews recommends audiological assessment and a medical genetics or genetic counselling consultation [11]. The ACMG recommends referral to a multidisciplinary centre with an otorhinolaryngologist, geneticist, genetic counsellor, audiologist, speech and language therapist and early intervention and family support services, including deaf people or parents of deaf children, and, given a syndromic diagnosis, a short- and long-term screening and management plan according to the syndrome [88]. In SOX10 forms, the initial assessment adds CT and MRI of the temporal bone before any implant decision, neurological and developmental assessment in neurological forms and investigation of the gonadotropic axis when there is anosmia or absent puberty [9, 55].

7.2. Auditory rehabilitation with hearing aids

Management of hearing loss depends on severity [11]. Hearing aids require specialist audiological assessment and replacement of earmoulds as the ear canal grows [57]. The JCIH recommends that fitting include measurement of the real-ear-to-coupler difference (RECD) with the earmould in place, to adjust amplification to individual targets across the speech spectrum, and that regular audiological assessment be performed to detect fluctuation or deterioration and reprogramme the hearing aids; validation of the fitting is mandatory [106]. Among the quality indicators, the JCIH includes the proportion of deaf or hard-of-hearing infants who receive amplification within the first month after confirmation, in families who choose it [106]. Education is adjusted to the degree of hearing loss [57], and communication in sign language, spoken language or both is achievable when the identification and intervention timelines are met [106].

7.3. Cochlear implant

According to the JCIH, when the parents' goals include spoken language, cochlear implantation is the main treatment for most children with severe to profound hearing loss; it is indicated above 12 months of age in severe to profound bilateral sensorineural hearing loss (including auditory neuropathy) without expected progress with adequate amplification, and the document allows for lowering the age approved by the US Food and Drug Administration (FDA) to under 12 months [106]. Surgery at around 12 months or earlier offers the greatest likelihood of open-set speech understanding and of language acquisition at the pace of hearing peers, and results are better with implantation by 2 years [106]. The Oxford Desk Reference considers implantation established in very young children with profound hearing loss and most appropriate in preschool children with a loss greater than 65 dB despite optimised hearing aids [57].

In WS, cochlear implantation has favourable results in several series [11, 115–118]. The meta-analysis by Vanstrum and colleagues brought together 20 articles, 192 patients and 210 implants: mean age at implantation of 3.8 years (95% confidence interval, 95% CI, 3.1–4.5), mean follow-up of 5.2 years (95% CI 3.4–7.0), surgical complications in 11 of 210 (5.2%), the most common being gusher (outflow of cerebrospinal fluid or perilymph), and favourable auditory outcome in 90% (95% CI 84–94%), also in the presence of temporal bone anomalies (p = 0.04) [119]. The “favourable outcome” variable was constructed from the description by the authors of the included studies, which limits comparison [119]. GeneGuidance cites a PRISMA systematic review of 16 studies with 179 patients and at least 194 implants, all of level III–IV of the Oxford classification (OCEBM), mainly uncontrolled case series, and a retrospective cohort of 14 children implanted by 2 years of age with speech perception and language comprehension similar to those of a reference group of deaf children without other disabilities [55].

In a Japanese series of 12 children with WS implanted between 1995 and 2020, with a mean age of 2.5 years at implantation, early intervention correlated with better scores on the 67S and CI-2004 tests; children without cochlear malformation or treatment delay had mean speech recognition above 80%, and all attended mainstream classes in lower secondary school [120]. The Chinese twins with a WS1 phenotype and a de novo MITF variant achieved, after implantation, a mean pure-tone threshold of 30 dB in free field [42].

7.4. Cochlear implantation in inner ear malformations and SOX10 forms

In SOX10 forms, inner ear malformation is the rule and absence of the cochlear nerve changes implant candidacy, so CT and MRI of the temporal bone should precede planning [55, 92]. The JCIH points out that cochlear nerve deficiency or severe inner ear malformations may directly affect the prognosis for speech understanding and spoken language with amplification or implantation, and that MRI is usually sufficient for implantation, reserving CT for facial nerve anomalies, inner ear ossification, patency of the bony cochlear nerve canal or temporal bone pathology [106]. In the Korean cohort of Shin and colleagues, SOX10 WS had the greatest reduction in cochlear duct length, frequently with vestibular malformations; the authors recommend preoperative imaging and genetic testing to choose the electrode [93]. Gusher was the most frequent complication in the meta-analysis [119].

7.5. Hirschsprung disease in WS4

Hirschsprung disease in WS4 follows the general management of Hirschsprung disease. Definitive treatment is resection of the aganglionic segment with pull-through surgery [55]. The 2020 ERNICA guideline for the rectosigmoid form covers histological diagnosis before surgery, preoperative screening for renal anomalies and malnutrition, postoperative care, management of associated enterocolitis and follow-up into adulthood with transition [15, 55]. In syndromic Hirschsprung disease, the guideline recommends referral to a genetics clinic and testing of the gene specific to the syndromic phenotype (level III, conditional recommendation in favour, 100% agreement), and cites Shah-Waardenburg syndrome among the associated syndromes [15]. In 4–30% of patients with Hirschsprung disease there is an associated syndrome or developmental disorder and in 15–20% aganglionosis extends beyond the rectosigmoid; postoperative enterocolitis occurs in about 30%, with family history and long segment among the risk factors [15]. In 2026 the update of the ERNICA guideline was published, prepared by 28 paediatric surgeons and a dietitian from 25 centres in 12 European countries, with four representatives of three patient associations, based on the evidence from 2018–2024 and using GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology [121]. The lowered threshold for rectal biopsy in associated syndromes and the recommendation of a genetics consultation with testing of the specific gene in syndromic Hirschsprung disease are maintained (level III, conditional in favour) [121]. Among the changes, the combination of haematoxylin-eosin and calretinin is recommended to improve diagnostic accuracy, especially in preterm infants, as well as screening renal ultrasound for congenital anomalies of the kidney and urinary tract in the first year of life (level II, conditional, new), preoperative screening for malnutrition (level II, strong, new) and attention to sexual health, fertility and transition to adult care; the panel notes cardiac comorbidities in 6–10% of patients [121].

The EUPSA survey of 294 members from 61 countries showed a preference for the Soave technique in 65% (mainly transanal), followed by the Duhamel and Swenson techniques; while awaiting surgery, rectal irrigations, some dilatations or a stoma are used, and persistent symptoms after pull-through are treated with enemas and laxatives; the survey itself concludes that most aspects of management lack consensus [111]. In patients with a syndrome and cognitive impairment, continence is frequently achieved later [15].

7.6. Neurological forms and the gonadotropic axis

In neurological forms (PCWH and neurological WS2E), management includes neurological and developmental assessment, nerve conduction studies, brain MRI and support for delay, spasticity and ataxia [55]. Chronic intestinal pseudo-obstruction and hypoganglionosis require intestinal surveillance even without classic Hirschsprung disease [9]. Anosmia and absent puberty are investigated and treated as isolated GnRH deficiency. In the two cases of Yang and colleagues, hypogonadotropic hypogonadism was treated with gonadotropins for virilisation and fertility preservation; the doses are not stated in the abstract and are not reproduced [104].

7.7. Skin, eyes and locomotor system

Pigmentary anomalies have no medical treatment; depigmented skin requires sun protection [55]. Visual acuity is normal in WS1 [11], but in the child with severe hearing loss the Oxford Desk Reference recommends a complete ophthalmological assessment, because the child depends on vision to communicate [57]. In WS3, management is orthopaedic, directed at contractures, carpal fusion and syndactyly [55].

7.8. Drugs and regulatory status

There is no medicine approved by the FDA or the European Medicines Agency (EMA) for any Waardenburg syndrome or for PCWH, nor any therapy directed at the cause [108, 114]. The only pharmacological intervention mentioned by GeneReviews is periconceptional folic acid supplementation in women at increased risk of having a child with WS1, because of the possible increase in the risk of neural tube defects; this recommendation is based on a murine study of embryonic folate metabolism, and no human study has defined the optimal dose [11, 122]. The recommendation is not extrapolated to the MITF forms [55]. Gonadotropins in SOX10 hypogonadism are used according to GnRH deficiency protocols [55, 104].

7.9. Clinical trials and outcome studies

The ClinicalTrials.gov search (condition “Waardenburg”, 10 Oct 2026) returned two studies, neither therapeutic: NCT02418936, of a genetic hearing loss diagnostic kit (Xiangya, observational, 100 participants, 2015–2017, status unknown), and NCT03655223, the Early Check expanded newborn screening study (RTI International, observational, 30,000 participants, from 15 Oct 2018 to 31 Dec 2026, active not recruiting) [108]. A negative search does not demonstrate the absence of preclinical research. The ACMG states that gene therapies for hearing loss were at the preclinical stage in 2022 [88]. The evidence on auditory rehabilitation comes from observational studies.

The studies are not comparable with each other: they differ in age at implantation, proportion of malformations, genes and outcome scales, and none has a randomised control group. The 90% proportion of favourable outcomes describes patients selected for implantation and published, not all patients with WS, and the influence of the gene on the outcome is poorly quantified.

7.10. Developments in the last 6–12 months

In the same period, there were no authorisations, regulatory opinions or registered therapeutic trials for Waardenburg syndromes in the sources searched [108].

7.11. Agents to avoid, pregnancy, lactation, perioperative care and transition

The sources consulted do not identify contraindicated drugs specific to Waardenburg syndromes. In the pregnancy of a woman with WS1, or of a couple in which one partner has WS1, periconceptional folic acid is recommended, without a defined human dose for this indication [11]. In the absence of such a dose, that used for the prevention of neural tube defects is applied by extrapolation: the USPSTF recommends 0.4–0.8 mg daily for all persons planning or capable of pregnancy (grade A) [124], and the Oxford Desk Reference indicates 400 μg daily from at least one month before conception until the 12th week to prevent first occurrence, reserving 5 mg daily, over the same period, for the prevention of recurrence after a child with a neural tube defect [125]. None of the sources defines a specific dose for the risk associated with WS1. The sources contain no specific recommendations for lactation. Perioperatively, the risk of gusher in cochlear implantation justifies preoperative imaging [93, 119], and the JCIH points out that general anaesthesia for repeated ABR in young children requires weighing of benefits and risks to cognitive function [106]. In Hirschsprung disease, preoperative screening for renal anomalies and malnutrition is part of the ERNICA guideline [15, 55]. Transition to adult care is provided for in the ERNICA guideline for Hirschsprung disease [15], and the ACMG recommends, in the absence of an aetiological diagnosis, genetic reassessment every 3 years, also in adulthood, with adapted assessment [88].

7.12. Guideline recommendations and divergences

There is no guideline specific to Waardenburg syndromes; the applicable recommendations come from documents on childhood hearing loss, Hirschsprung disease and gene-disease validity.

There are three divergences. The first concerns audiological surveillance: GeneReviews considers repeating the audiogram usually unnecessary in WS1, because the hearing loss does not progress [11], whereas the JCIH requires regular audiological assessment of children with hearing aids to detect fluctuation or deterioration [106]; the GeneReviews guidance does not apply to the neurological SOX10 forms [55]. The second concerns the frequency of hearing loss used in counselling: “about 50%” in the ACMG versus 71.0% in the systematic review [13, 88]. The third concerns the W index: GeneReviews retains the 1.95 threshold, and Japanese and American studies question its validity and propose raising the cut-off [44, 89].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks by gene and scenario

8.2. WS1: counselling of the proband and family

Most people with WS1 have an affected parent, and a minority have a de novo variant [11]. In the parents of a simplex case, examination of the face with calculation of the W index, examination of the skin and hair and an audiogram are recommended, with targeted molecular testing if the variant is known [11]. If the variant is not found in the parents and parentage is confirmed, the explanation is a de novo variant or germline mosaicism, or somatic and germline mosaicism; germline mosaicism has been documented since 1991 and confirmed molecularly for PAX3 in 2018 [11, 126, 127]. An apparently negative family history may reflect a mild phenotype in a parent [11].

The risk of hearing loss in offspring is calculated by multiplying the probability of transmission (50%) by the probability of expression of hearing loss [128]. The Smith teaching example uses 20% expression and obtains 10% [128]; with current frequencies of hearing loss in WS1 (52.3% in the systematic review, 47–58% in clinical series), the risk of a child of a parent with WS1 having hearing loss is about 24–29% [11, 13]. The calculation is an illustration: it does not distinguish degrees of hearing loss, and intrafamilial variability prevents individual predictions [11]. Absence of expression in a carrier parent, which reflects variability, must be distinguished from a de novo variant [128]. SGRT presents internally inconsistent figures (about 90% with affected relatives and “only 25% with a new mutation”) and a “penetrance of 36–58%” that does not correspond to the almost complete penetrance described in GeneReviews [11, 56].

8.3. Gene-specific features

In MITF carriers, the risk of clinically significant hearing loss is about one third, and the risk of Hirschsprung disease is small or nil [57]. In SOX10 forms, most variants were initially de novo; familial cases now account for about 20%, with great intra- and interfamilial variability, and parental mosaicism has been found in about 3–4% (a higher proportion in a small series with more sensitive methods) [9]. The risk of PCWH after a first child without PCWH is considered low, and the birth of a child with WS4 after a first, milder case with WS2 has been described several times, without the real risk being quantifiable owing to publication bias [9]. The Chinese family with SOX10 p.(Arg43Ter) shows that a carrier parent may have no signs [79]. In EDN3 and EDNRB forms, complete disease usually requires two alleles, and heterozygotes may have isolated Hirschsprung disease [12, 55]. The implication for couples in which both partners have WS due to the same gene (PAX3, MITF or SOX10) follows from the inheritance: a 25% probability of a severe biallelic form (WS3 or Klein syndrome for PAX3, COMMAD for MITF, polymalformative fetal phenotype for SOX10); the sources provide no empirical estimates [9, 68, 70].

8.4. Cascade testing and at-risk relatives

It is appropriate to assess at-risk relatives to allow early screening for hearing loss: molecular testing if the familial variant is known; otherwise, clinical examination for WS1 features and audiological assessment [11]. In SOX10 forms, at-risk relatives are assessed by targeted testing or by examination and audiology [55]. For KITLG, incomplete penetrance means that a carrier relative with normal hearing and skin is expected, and a normal audiogram does not exclude the genotype [55].

8.5. Prenatal diagnosis and preimplantation genetic testing

Prenatal diagnosis and preimplantation genetic testing are possible once the PAX3 variant (or that of another gene) has been identified in the family, but they do not predict the manifestations or severity; perspectives on their use differ, and in most centres the decision is personal [11]. The Oxford Desk Reference considers prenatal diagnosis of hearing loss “highly controversial”, technically possible by chorionic villus sampling when the variants are known [57]. GeneReviews recommends determining the risk and discussing prenatal and preimplantation diagnosis before pregnancy, counselling affected or at-risk young adults and storing DNA from probands without a molecular diagnosis [11, 129]. Fetal ultrasound may detect neural tube defects, as in the prenatal diagnosis of spina bifida in a family with WS1 [98] and in the case of Chen and colleagues, in which the PAX3 deletion was identified in fetal tissue from a pregnancy with spina bifida and then in the amniotic fluid of the following pregnancy, whose newborn had a normal phenotype at birth [100].

8.6. Communication, Deaf cultural perspective and minors

The ACMG recommends offering genetic evaluation and counselling to all children with confirmed hearing loss, with sign language interpreters trained in genetic terminology, video calls, written messages and visual resources [88]. The guideline asks for respect for the Deaf cultural perspective, which sees deafness as a neutral or positive linguistic and cultural trait: using “chance” instead of “risk” and “deaf” or “hearing” instead of “affected” or “unaffected”, and avoiding terms such as “disability” or “pathology” [88]. When one of the parents has profound congenital deafness, a sign language interpreter should be provided at the consultation [57]. Pre-test counselling includes the decision not to test and, for exome or genome, discussion of secondary findings; post-test counselling addresses the stability or progression of hearing loss, referrals, support organisations, inheritance and the chance for future children [88]. Testing of minors is justified when it guides auditory surveillance or early management; reproductive counselling is directed at young adults [11].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

The Dutch value of 2.4/100,000 is equivalent to about 1 in 42,000 and is consistent with the original estimate. The European value from congenital anomaly registries is about six times lower, because many cases are not recognised at birth; GeneGuidance transcribed it as “0.37 per 1,000,000”, a unit error that the Orphanet record corrects [1, 55]. The values of 1/40,000 and 2–5% of congenital deafness circulating in the literature are citations without a primary population study having been read [53, 114]. The class “<1/1,000,000” and the record of “3 families” for WS2 in Orphanet reflect the count of families in two articles and are incompatible with the clinical frequency of WS2, which is one of the most common types [3, 105]; they should not be read as prevalence. Hirschsprung disease, from all causes, affects about 1 in 5000 newborns, with higher frequency in Asians, intermediate in Caucasians and lower in African Americans [9, 54].

9.2. USA and Europe

American data include the Waardenburg Consortium families [22], the Mennonite population of Pennsylvania, in which EDNRB p.(Trp276Cys) is a founder allele responsible for a significant proportion of cases of Hirschsprung disease [12, 29], and the Children's Hospital of Philadelphia cohort on the W index [89]. In Europe, data from the United Kingdom (Manchester) underpinned the feature frequencies and the WS2 criteria [23, 64]; French data underpin the SOX10 spectrum and temporal bone imaging [9, 92] and the national KITLG series [55]; Sweden contributed enteric histopathology [81] and Ireland SOX10 neuropathy [103]. In Turkey, a consanguineous family documented PAX3 homozygosity in WS3, and four other families with WS1 had two missense variants, one splicing variant and one small insertion [68]. The sources consulted contain no data from Portugal; this absence does not demonstrate that no published data exist.

9.3. Brazil and Latin America

Brazil has made central contributions. The Recife family studied by da-Silva, with 73 affected individuals in two pedigrees, was at the origin of the identification of PAX3 [25, 26]. Pardono and colleagues, in São Paulo, established the clinical differentiation between WS1 and WS2 and the superiority of the family mean W index [85]. Bocángel and colleagues studied 49 probands, mostly from the Southeast, with 38.8% molecular diagnosis, a relevant weight of deletions detected by MLPA and at least 10 novel variants [82]. SGRT, a Brazilian work, presents a mother and son from Fortaleza with WS1 and cites Brazilian studies on audiology, ophthalmology, dermatology and dental phenotype, without national prevalence data [56]. In Colombia, the screening programme of Tamayo and colleagues identified 95 affected individuals in the institutionalised deaf population [86].

9.4. Asia and Africa

In Japan, the study of 16 families showed that the W index is not valid in all ethnic groups [44], and the implant series of Arai and colleagues documented long-term outcomes [120]. In China, more than 60 PAX3 variants have been described, with about 48% truncating and possible hotspots (c.667C>T, c.117C>A, c.123del, c.127G>T, c.452-2A>G), according to the review by Zhan and colleagues, which does not state the transcript of these variants [65]; Yang and colleagues described the heterogeneity of WS2 [84]. In Hong Kong, congenital sensorineural hearing loss was less frequent than in other Chinese cohorts [109]. In Korea, the implant cohort of Shin and colleagues documented the short cochlear duct in SOX10 WS [93]. In India, Shah described the form with long-segment Hirschsprung disease [19]. The African systematic review by Aboagye and colleagues brought together 15 articles and 84 cases, without sex bias: congenital profound sensorineural hearing loss in 56 (66.7%), WS2 in 31 (36.9%), pathogenic variants in PAX3 (13 families), SOX10 (7), EDNRB (4) and EDN3 (1) in Morocco, Tunisia and South Africa, most cases in South Africa (32; 38.1%) and Tunisia (22; 26.2%), PAX8 as a candidate in a family from Ghana and variants in non-syndromic hearing loss genes (BDP1, MYO6) in two South African families; of the 43 variant types, 27 were missense, 7 deletions, 5 splicing, 2 nonsense, 1 indel and 1 duplication, without functional studies [105].

9.5. Special populations

The white forelock may be hidden by hair dye, appear only in adolescence or repigment, so the absence of a forelock on examination does not exclude the diagnosis [11]. The W index threshold was derived in specific populations and fails in some ethnic groups [44, 89]. In children with profound deafness and deaf parents, sign language communication and the Deaf cultural perspective shape counselling [57, 88]. In children with a syndrome and cognitive impairment after surgery for Hirschsprung disease, continence is frequently achieved later [15].

9.6. European reference networks and patient organisations

ERN CRANIO (European network for craniofacial anomalies and rare ear, nose and throat disorders) includes genetic hearing loss among the ENT disorders it covers, with 35 hospitals in 15 countries [14]. ERNICA covers Hirschsprung disease and drew up the 2020 and 2025 guidelines [15, 121]. The sources consulted identify no other European network specific to the pigmentary or ocular manifestations.

In the USA, GeneReviews and the National Organization for Rare Disorders (NORD) list MedlinePlus, the American Society for Deaf Children, BabyHearing.org of the National Institute on Deafness and Other Communication Disorders (NIDCD) (newborn hearing screening), the Hereditary Hearing Loss Homepage, the National Association of the Deaf, NOAH (National Organization for Albinism and Hypopigmentation), FACES, the Hearing Loss Association of America, the American Academy of Audiology, the Let Them Hear Foundation and the National Center on Deaf-Blindness [11, 114]. In the United Kingdom, the Oxford Desk Reference lists the National Deaf Children's Society, Action on Hearing Loss and DeafPLUS, and Contact reports that there is no WS-specific support group and maintains a general helpline [57, 130]. In Hirschsprung disease, the associations SoMA (Germany) and A.Mor.Hi (Italy) took part in the ERNICA guideline [15]. Reference variant databases include the Deafness Variation Database, the LOVD (Leiden Open Variation Database) for PAX3, HGMD and ClinVar [11].

10. Illustrative Clinical Case

10.1. Presentation

In 2025 Zhan and colleagues described a 17-year-old girl from a Yugur family of Zhangye, in Gansu province (China), with congenital bilateral deafness [65]. Pregnancy and delivery were normal, but the child failed newborn hearing screening. The parents, non-consanguineous, had normal hearing and no other signs, and no other relative had similar symptoms; the pedigree shows a sporadic case (Figure 1A) [65]. On examination there was a broad nasal root, iris heterochromia and a thicker left eyebrow. The text of the article describes a blue left iris and a normally pigmented right iris, whereas the legend of Figure 1 states “brown on the left, blue on the right”; fundus photography shows a right ocular fundus with uniform pigmentation and a left fundus with markedly reduced pigmentation (Figure 2A) [65]. The inner intercanthal, interpupillary and outer intercanthal distances were 40, 66 and 95 mm, with a W index of about 2.10, above the 1.95 threshold [65].

10.2. Investigation

Preoperative pure-tone audiometry showed profound bilateral sensorineural hearing loss at all frequencies (>90 dB HL) (Figure 1C and 1D), and ABR evoked no waves. CT of the temporal bone and MRI of the inner ear showed no malformations. Optical coherence tomography showed normal foveal architecture and intact retinal layers in both eyes (Figure 2B), with a mean nerve fibre layer thickness of 98 µm on the right and 100 µm on the left and cup-to-disc ratios of 0.37 and 0.31; visual acuity was normal, without colour vision or visual field deficits. There were no musculoskeletal anomalies, Hirschsprung disease or patchy skin depigmentation. The clinical diagnosis of WS1 was based on the Waardenburg Consortium criteria [65].

Exome sequencing (mean depth of 203.59×, with 99.49% of bases above 20×) and Sanger confirmation in the trio identified in PAX3 the heterozygous variant NM_181458.4:c.788dup p.(Gln264ThrfsTer5) (NC_000002.12:g.222232082dup), de novo, in exon 5, in the homeodomain region, absent from the HGMD, gnomAD and ExAC databases; with the criteria PVS1 and PM2_supporting, it was classified as likely pathogenic (Figure 3, sequence of the proband and parents and conservation of residue Gln264, which the legend designates “(C)788dup”) [58, 65]. The exome also revealed VUS in NLRP3, DNMT1 and GJB3, not considered causal [65].

10.3. Management and course

At 6 years of age, the child received a unilateral right cochlear implant (Cochlear CI24RE (CA)); surgery showed no ossicular or cochlear malformations, ossification or outflow of perilymph or cerebrospinal fluid [65]. Before implantation she had no perception of environmental sounds or speech and responded only to gestures. Three years later she moved to mainstream education, and 7 years after implantation she communicated by telephone with familiar people [65].

CAP: Categories of Auditory Performance; SIR: Speech Intelligibility Rating; MAIS: Meaningful Auditory Integration Scale; MUSS: Meaningful Use of Speech Scale. Source: Zhan 2025, Table 1 [65].

10.4. Interpretation

The case brings together the elements supporting the diagnosis of WS1: two major criteria (congenital hearing loss and pigmentary anomaly of the iris) and a W index above the threshold, confirmed by a truncating PAX3 variant in the homeodomain [11, 65]. The absence of a family history is explained by the de novo variant, a situation that, according to GeneReviews, is a minority in WS1 [11]; the risk for future siblings is slightly higher than the population risk owing to the possibility of germline mosaicism, and the risk for the patient's offspring is 50% [11, 127]. The unilateral hypopigmentation of the ocular fundus with normal acuity corresponds to the sectoral pattern described by Shields and colleagues [52]. Implantation at 6 years is late for a congenital profound hearing loss detected on newborn screening; the JCIH recommends surgery at around 12 months or earlier, with better results by 2 years [106], and the good outcome in this case does not justify delaying intervention. The absence of inner ear malformations, expected in PAX3 WS1 and different from the pattern of the SOX10 forms, favoured uncomplicated surgery [9, 65].

Two complementary cases illustrate the extremes of the PAX3 spectrum. In the Turkish family of Wollnik and colleagues, the parents heterozygous for p.(Tyr90His) had WS1 and the homozygous son, at 18 months, WS3 with upper limb defects [68]. In the family of Chen and colleagues, the same paternal deletion of exons 1 to 4 was associated with spina bifida in one pregnancy and a normal phenotype at birth in the next [100].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

The sources consulted do not describe mortality specific to WS1, WS2 or WS3; functional prognosis depends on the hearing loss and on how early rehabilitation is started [11, 56]. With early detection and appropriate therapy, neuropsychological development is normal [56], and in the implant meta-analysis the auditory outcome was favourable in 90% (95% CI 84–94%) [119]. Hearing loss in WS1 does not progress [11]. In WS4, morbidity and mortality result mainly from enterocolitis [56], which occurs in about 30% after surgery [15]. In PCWH, prognosis depends on the severity of the myelin defects, with pictures ranging from mild delay to forms with cerebral palsy, and some SOX10 neurological phenotypes are progressive [9, 54]. WS1 does not increase oncological risk, because the PAX3-FOXO1 fusion of rhabdomyosarcoma is somatic [11]. SOX10 expression in melanoma, breast cancer, gliomas and other tumours is tumoral, with no demonstrated germline risk [9].

In carriers of MITF p.Glu318Lys, who do not have WS, in a 14-year hospital dermatological surveillance none of the 12 carriers without previous cancer developed melanoma during a mean follow-up of 24.6 months (2 basal cell carcinomas were detected) [55, 123]. A meta-analysis found no evidence of a relevant risk of non-melanoma cancers (odds ratio 1.168; 95% CI 0.78–1.74; p = 0.454) [55, 131]. Dermatological surveillance of these carriers includes dermoscopy, self-examination and photoprotection, with total body photography when there are many naevi; renal surveillance is not established [55].

11.2. Surveillance

11.3. Follow-up in adulthood

Transition to adult care is structured in the ERNICA guideline for Hirschsprung disease [15]. In adulthood, genetic follow-up includes reproductive counselling, review of variant classification and, in families without a molecular diagnosis, periodic reassessment [11, 88]. In SOX10 forms, hypogonadism may be recognised only at puberty or later, and treatment with gonadotropins aims at virilisation and fertility [9, 104].

12. Summary and Final Considerations

Waardenburg syndromes result from the lack of functional melanocytes in the skin, hair, eyes and stria vascularis, and share congenital sensorineural hearing loss and pigmentary anomalies. Type 1 is defined by dystopia canthorum and is due in more than 90% of cases to PAX3 variants detectable by sequencing, with about 6% deletions; type 2 is genetically heterogeneous (MITF, SOX10, heterozygous EDNRB and, with limited evidence, KITLG and SNAI2); type 3 adds limb anomalies and frequently corresponds to PAX3 homozygosity; type 4 adds Hirschsprung disease and is due to heterozygous SOX10 or to EDN3 and EDNRB, usually biallelic; PCWH, due to SOX10 variants that escape NMD, adds peripheral neuropathy and leukodystrophy. Hearing loss occurs in about half of patients with WS1 and in more than 80% of patients with WS2 and WS4, and the gene predicts hearing better than the clinical type.

Diagnosis combines the Waardenburg Consortium criteria, measurement of the W index in the family, whose 1.95 threshold fails in some populations, and a panel with copy-number analysis, because PAX3 and SOX10 deletions escape sequencing. There is no disease-modifying therapy and no registered therapeutic trial. Management rests on newborn hearing screening with the 1-3-6 goals, auditory rehabilitation with hearing aids and early cochlear implantation, with a favourable outcome in about 90% of implant recipients, temporal bone imaging before implantation in SOX10 forms, rectal biopsy with a lowered threshold and surgery for Hirschsprung disease according to ERNICA, and neurological and pubertal surveillance of SOX10 forms. Counselling starts from dominant inheritance with unpredictable expressivity in WS1, considers germline and parental mosaicism, the dosage effect of the endothelin pathway and the risk of severe biallelic forms when both parents are carriers, and respects the Deaf cultural perspective.

13. References

1. Orphanet. Waardenburg syndrome (ORPHA:3440). Paris: INSERM; [accessed 2026-10-10]. Available at: https://www.orpha.net/en/disease/detail/3440
2. Orphanet. Waardenburg syndrome type 1 (ORPHA:894). [accessed 2026-10-10]. https://www.orpha.net/en/disease/detail/894
3. Orphanet. Waardenburg syndrome type 2 (ORPHA:895). [accessed 2026-10-10]. https://www.orpha.net/en/disease/detail/895
4. Orphanet. Waardenburg syndrome type 3 (ORPHA:896). [accessed 2026-10-10]. https://www.orpha.net/en/disease/detail/896
5. Orphanet. Waardenburg-Shah syndrome (ORPHA:897). [accessed 2026-10-10]. https://www.orpha.net/en/disease/detail/897
6. Orphanet. Peripheral demyelinating neuropathy-central dysmyelinating leukodystrophy-Waardenburg syndrome-Hirschsprung disease (ORPHA:163746). [accessed 2026-10-10]. https://www.orpha.net/en/disease/detail/163746
7. Orphanet. Tietz syndrome (ORPHA:42665). [accessed 2026-10-10]. https://www.orpha.net/en/disease/detail/42665
8. Salah S, Meiner V, Abumayaleh A, Asafra A, Al-Sharif T, Al-Fallah O, et al. Biallelic variants in PAX3 cause Klein syndrome. Clin Genet. 2022;102(3):223-227. doi:10.1111/cge.14167. PMID: 35607853.
9. Pingault V, Zerad L, Bertani-Torres W, Bondurand N. SOX10: 20 years of phenotypic plurality and current understanding of its developmental function. J Med Genet. 2022;59(2):105-114. doi:10.1136/jmedgenet-2021-108105. PMID: 34667088. PMC8788258.
10. Online Mendelian Inheritance in Man, OMIM®. Johns Hopkins University, Baltimore (MD). MIM #193510 — Waardenburg syndrome, type 2A; WS2A [Internet]. [accessed 2026-10-10]. Available at: https://www.omim.org/entry/193510
11. Milunsky JM. Waardenburg Syndrome Type I. 2001 Jul 30 [updated 2022 Oct 20]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. NBK1531. PMID: 20301703. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1531/
12. Parisi MA. Hirschsprung Disease Overview – RETIRED CHAPTER, FOR HISTORICAL REFERENCE ONLY. 2002 Jul 12. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. NBK1439. PMID: 20301612. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1439/
13. Song J, Feng Y, Acke FR, Coucke P, Vleminckx K, Dhooge IJ. Hearing loss in Waardenburg syndrome: a systematic review. Clin Genet. 2016;89(4):416-425. doi:10.1111/cge.12631. PMID: 26100139.
14. ERN CRANIO — European Reference Network for rare and/or complex craniofacial anomalies and ear, nose and throat (ENT) disorders. Ear, Nose and Throat (ENT) disorders [Internet]. [accessed 2026-10-10]. Available at: https://www.ern-cranio.eu/ear-nose-and-throat-ent-disorders
15. Kyrklund K, Sloots CEJ, de Blaauw I, Bjørnland K, Rolle U, Cavalieri D, et al. ERNICA guidelines for the management of rectosigmoid Hirschsprung's disease. Orphanet J Rare Dis. 2020;15(1):164. doi:10.1186/s13023-020-01362-3. PMID: 32586397. PMC7318734.
16. Waardenburg PJ. A new syndrome combining developmental anomalies of the eyelids, eyebrows and nose root with pigmentary defects of the iris and head hair and with congenital deafness. Am J Hum Genet. 1951;3(3):195-253. PMID: 14902764. PMC1716407.
17. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Waardenburg syndrome; p. 358–361.
18. Arias S, Mota M. Apparent non-penetrance for dystopia in Waardenburg syndrome type I, with some hints on the diagnosis of dystopia canthorum. J Genet Hum. 1978;26(2):103-31. PMID: 670940.
19. Shah KN, Dalal SJ, Desai MP, Sheth PN, Joshi NC, Ambani LM. White forelock, pigmentary disorder of irides, and long segment Hirschsprung disease: possible variant of Waardenburg syndrome. J Pediatr. 1981;99(3):432-5. doi:10.1016/s0022-3476(81)80339-3. PMID: 7264803.
20. Online Mendelian Inheritance in Man, OMIM®. Johns Hopkins University, Baltimore (MD). MIM #148820 — Waardenburg syndrome, type 3; WS3 [Internet]. [accessed 2026-10-10]. Available at: https://www.omim.org/entry/148820
21. Sheffer R, Zlotogora J. Autosomal dominant inheritance of Klein-Waardenburg syndrome. Am J Med Genet. 1992;42(3):320-2. doi:10.1002/ajmg.1320420312. PMID: 1536170.
22. Farrer LA, Grundfast KM, Amos J, Arnos KS, Asher JH Jr, Beighton P, et al. Waardenburg syndrome (WS) type I is caused by defects at multiple loci, one of which is near ALPP on chromosome 2: first report of the WS consortium. Am J Hum Genet. 1992;50(5):902-13. PMID: 1349198. PMC1682585.
23. Liu XZ, Newton VE, Read AP. Waardenburg syndrome type II: phenotypic findings and diagnostic criteria. Am J Med Genet. 1995;55(1):95-100. doi:10.1002/ajmg.1320550123. PMID: 7702105.
24. Tassabehji M, Read AP, Newton VE, Harris R, Balling R, Gruss P, et al. Waardenburg's syndrome patients have mutations in the human homologue of the Pax-3 paired box gene. Nature. 1992;355(6361):635-6. doi:10.1038/355635a0. PMID: 1347148.
25. Baldwin CT, Hoth CF, Amos JA, da-Silva EO, Milunsky A. An exonic mutation in the HuP2 paired domain gene causes Waardenburg's syndrome. Nature. 1992;355(6361):637-8. doi:10.1038/355637a0. PMID: 1347149.
26. da-Silva EO. Waardenburg I syndrome: a clinical and genetic study of two large Brazilian kindreds, and literature review. Am J Med Genet. 1991;40(1):65-74. doi:10.1002/ajmg.1320400113. PMID: 1887852.
27. Hoth CF, Milunsky A, Lipsky N, Sheffer R, Clarren SK, Baldwin CT. Mutations in the paired domain of the human PAX3 gene cause Klein-Waardenburg syndrome (WS-III) as well as Waardenburg syndrome type I (WS-I). Am J Hum Genet. 1993;52(3):455-62. PMID: 8447316. PMC1682157.
28. Tassabehji M, Newton VE, Read AP. Waardenburg syndrome type 2 caused by mutations in the human microphthalmia (MITF) gene. Nat Genet. 1994;8(3):251-5. doi:10.1038/ng1194-251. PMID: 7874167.
29. Puffenberger EG, Hosoda K, Washington SS, Nakao K, deWit D, Yanagisawa M, et al. A missense mutation of the endothelin-B receptor gene in multigenic Hirschsprung's disease. Cell. 1994;79(7):1257-66. doi:10.1016/0092-8674(94)90016-7. PMID: 8001158.
30. Edery P, Attié T, Amiel J, Pelet A, Eng C, Hofstra RM, et al. Mutation of the endothelin-3 gene in the Waardenburg-Hirschsprung disease (Shah-Waardenburg syndrome). Nat Genet. 1996;12(4):442-4. doi:10.1038/ng0496-442. PMID: 8630502.
31. Hofstra RM, Osinga J, Tan-Sindhunata G, Wu Y, Kamsteeg EJ, Stulp RP, et al. A homozygous mutation in the endothelin-3 gene associated with a combined Waardenburg type 2 and Hirschsprung phenotype (Shah-Waardenburg syndrome). Nat Genet. 1996;12(4):445-7. doi:10.1038/ng0496-445. PMID: 8630503.
32. Pingault V, Bondurand N, Kuhlbrodt K, Goerich DE, Préhu MO, Puliti A, et al. SOX10 mutations in patients with Waardenburg-Hirschsprung disease. Nat Genet. 1998;18(2):171-3. doi:10.1038/ng0298-171. PMID: 9462749.
33. Inoue K, Tanabe Y, Lupski JR. Myelin deficiencies in both the central and the peripheral nervous systems associated with a SOX10 mutation. Ann Neurol. 1999;46(3):313-8. doi:10.1002/1531-8249(199909)46:3<313::aid-ana6>3.0.co;2-7. PMID: 10482261.
34. Inoue K, Khajavi M, Ohyama T, Hirabayashi S, Wilson J, Reggin JD, et al. Molecular mechanism for distinct neurological phenotypes conveyed by allelic truncating mutations. Nat Genet. 2004;36(4):361-9. doi:10.1038/ng1322. PMID: 15004559.
35. Sánchez-Martín M, Rodríguez-García A, Pérez-Losada J, Sagrera A, Read AP, Sánchez-García I. SLUG (SNAI2) deletions in patients with Waardenburg disease. Hum Mol Genet. 2002;11(25):3231-6. doi:10.1093/hmg/11.25.3231. PMID: 12444107.
36. Sánchez-Martín M, Pérez-Losada J, Rodríguez-García A, González-Sánchez B, Korf BR, Kuster W, et al. Deletion of the SLUG (SNAI2) gene results in human piebaldism. Am J Med Genet A. 2003;122A(2):125-32. doi:10.1002/ajmg.a.20345. PMID: 12955764.
37. Bondurand N, Dastot-Le Moal F, Stanchina L, Collot N, Baral V, Marlin S, et al. Deletions at the SOX10 gene locus cause Waardenburg syndrome types 2 and 4. Am J Hum Genet. 2007;81(6):1169-85. doi:10.1086/522090. PMID: 17999358. PMC2276340.
38. Pingault V, Bodereau V, Baral V, Marcos S, Watanabe Y, Chaoui A, et al. Loss-of-function mutations in SOX10 cause Kallmann syndrome with deafness. Am J Hum Genet. 2013;92(5):707-24. doi:10.1016/j.ajhg.2013.03.024. PMID: 23643381. PMC3644631.
39. Zazo Seco C, Serrão de Castro L, van Nierop JW, Morín M, Jhangiani S, Verver EJ, et al. Allelic Mutations of KITLG, Encoding KIT Ligand, Cause Asymmetric and Unilateral Hearing Loss and Waardenburg Syndrome Type 2. Am J Hum Genet. 2015;97(5):647-60. doi:10.1016/j.ajhg.2015.09.011. PMID: 26522471. PMC4667106.
40. Issa S, Bondurand N, Faubert E, Poisson S, Lecerf L, Nitschke P, et al. EDNRB mutations cause Waardenburg syndrome type II in the heterozygous state. Hum Mutat. 2017;38(5):581-593. doi:10.1002/humu.23206. PMID: 28236341.
41. Pingault V, Ente D, Dastot-Le Moal F, Goossens M, Marlin S, Bondurand N. Review and update of mutations causing Waardenburg syndrome. Hum Mutat. 2010;31(4):391-406. doi:10.1002/humu.21211. PMID: 20127975. Open-access author manuscript: HAL inserm-00483195, https://inserm.hal.science/inserm-00483195
42. Li W, Feng Y, Chen H, He C, Mei L, Liu XZ, et al. MITF Is Mutated in Type 1 Waardenburg Syndrome With Unusual Phenotype. Otol Neurotol. 2020;41(10):e1250-e1255. doi:10.1097/MAO.0000000000002821. PMID: 32740552.
43. Morimoto N, Mutai H, Namba K, Kaneko H, Kosaki R, Matsunaga T. Homozygous EDNRB mutation in a patient with Waardenburg syndrome type 1. Auris Nasus Larynx. 2018;45:222-6. doi:10.1016/j.anl.2017.03.022. PMID: 28502583.
44. Minami SB, Nara K, Mutai H, Morimoto N, Sakamoto H, Takiguchi T, et al. A clinical and genetic study of 16 Japanese families with Waardenburg syndrome. Gene. 2019;704:86-90. doi:10.1016/j.gene.2019.04.023. PMID: 30978479.
45. Verheij JB, Kunze J, Osinga J, van Essen AJ, Hofstra RM. ABCD syndrome is caused by a homozygous mutation in the EDNRB gene. Am J Med Genet. 2002;108(3):223-5. doi:10.1002/ajmg.10172. PMID: 11891690.
46. Online Mendelian Inheritance in Man, OMIM®. Johns Hopkins University, Baltimore (MD). MIM #277580 — Waardenburg syndrome, type 4A; WS4A [Internet]. [accessed 2026-10-10]. Available at: https://www.omim.org/entry/277580
47. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Microphthalmia and anophthalmia; p. 236–238.
48. Smith SD, Kelley PM, Kenyon JB, Hoover D. Tietz syndrome (hypopigmentation/deafness) caused by mutation of MITF. J Med Genet. 2000;37(6):446-8. doi:10.1136/jmg.37.6.446. PMID: 10851256. PMC1734605.
49. Read AP, Newton VE. Waardenburg syndrome. J Med Genet. 1997;34(8):656-65. doi:10.1136/jmg.34.8.656. PMID: 9279758. PMC1051028.
50. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Patchy hypo- or depigmented skin lesions; p. 276–277.
51. Merchant SN, McKenna MJ, Baldwin CT, Milunsky A, Nadol JB Jr. Otopathology in a case of type I Waardenburg's syndrome. Ann Otol Rhinol Laryngol. 2001;110:875-82. doi:10.1177/000348940111000913. PMID: 11558766.
52. Shields CL, Nickerson SJ, Al-Dahmash S, Shields JA. Waardenburg syndrome: iris and choroidal hypopigmentation: findings on anterior and posterior segment imaging. JAMA Ophthalmol. 2013;131:1167-73. doi:10.1001/jamaophthalmol.2013.4190. PMID: 23868078.
53. Bondurand N, Pingault V, Goerich DE, Lemort N, Sock E, Le Caignec C, et al. Interaction among SOX10, PAX3 and MITF, three genes altered in Waardenburg syndrome. Hum Mol Genet. 2000;9(13):1907-17. doi:10.1093/hmg/9.13.1907. PMID: 10942418.
54. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Hirschsprung's disease; p. 452–453.
55. GeneGuidance. Clinical gene curation: PAX3, MITF, SOX10, SNAI2, EDNRB, EDN3, KITLG [user database]. Accessed 10 Oct 2026.
56. Mororó GP, Mororó GP, Ribeiro EM. Síndrome de Waardenburg. In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 68–70.
57. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Deafness in early childhood; p. 122–125.
58. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166.
59. Corry GN, Underhill DA. Pax3 target gene recognition occurs through distinct modes that are differentially affected by disease-associated mutations. Pigment Cell Res. 2005;18:427-38. doi:10.1111/j.1600-0749.2005.00275.x. PMID: 16280008.
60. Corry GN, Hendzel MJ, Underhill DA. Subnuclear localization and mobility are key indicators of PAX3 dysfunction in Waardenburg syndrome. Hum Mol Genet. 2008;17:1825-37. doi:10.1093/hmg/ddn076. PMID: 18325909.
61. Birrane G, Soni A, Ladias JA. Structural basis for DNA recognition by the human PAX3 homeodomain. Biochemistry. 2009;48:1148-55. doi:10.1021/bi802052y. PMID: 19199574.
62. Wu TF, Yao YL, Lai IL, Lai CC, Lin PL, Yang WM. Loading of PAX3 to mitotic chromosomes is mediated by arginine methylation and associated with Waardenburg syndrome. J Biol Chem. 2015;290:20556-64. doi:10.1074/jbc.M114.607713. PMID: 26149688.
63. Milunsky JM, Maher TA, Ito M, Milunsky A. The value of MLPA in Waardenburg syndrome. Genet Test. 2007;11:179-82. doi:10.1089/gte.2006.0531. PMID: 17627390.
64. Tassabehji M, Newton VE, Liu XZ, Brady A, Donnai D, Krajewska-Walasek M, et al. The mutational spectrum in Waardenburg syndrome. Hum Mol Genet. 1995;4(11):2131-7. doi:10.1093/hmg/4.11.2131. PMID: 8589691.
65. Zhan L, Xu B, Lin D, Wang Y, Bian P. A novel frameshift variant of PAX3 in a Chinese Yugur family with Waardenburg syndrome type 1. Front Genet. 2025;16:1679351. doi:10.3389/fgene.2025.1679351. PMID: 41195004. PMC12585917.
66. DeStefano AL, Cupples LA, Arnos KS, Asher JH, Baldwin CT, Blanton S, et al. Correlation between Waardenburg syndrome phenotype and genotype in a population of individuals with identified PAX3 mutations. Hum Genet. 1998;102:499-506. doi:10.1007/s004390050732. PMID: 9654197.
67. Asher JH Jr, Sommer A, Morell R, Friedman TB. Missense mutation in the paired domain of PAX3 causes craniofacial-deafness-hand syndrome. Hum Mutat. 1996;7(1):30-5. doi:10.1002/(SICI)1098-1004(1996)7:1<30::AID-HUMU4>3.0.CO;2-T. PMID: 8664898.
68. Wollnik B, Tukel T, Uyguner O, Ghanbari A, Kayserili H, Emiroglu M, et al. Homozygous and heterozygous inheritance of PAX3 mutations causes different types of Waardenburg syndrome. Am J Med Genet A. 2003;122A(1):42-5. doi:10.1002/ajmg.a.20260. PMID: 12949970.
69. Grill C, Bergsteinsdóttir K, Ogmundsdóttir MH, Pogenberg V, Schepsky A, Wilmanns M, et al. MITF mutations associated with pigment deficiency syndromes and melanoma have different effects on protein function. Hum Mol Genet. 2013;22(21):4357-67. doi:10.1093/hmg/ddt285. PMID: 23787126. PMC3888191.
70. George A, Zand DJ, Hufnagel RB, Sharma R, Sergeev YV, Legare JM, et al. Biallelic Mutations in MITF Cause Coloboma, Osteopetrosis, Microphthalmia, Macrocephaly, Albinism, and Deafness. Am J Hum Genet. 2016;99(6):1388-1394. doi:10.1016/j.ajhg.2016.11.004. PMID: 27889061. PMC5142105.
71. Pampanos A, Kostantinidou AE, Tzetis M, Papanikolaou V, Falaina V, Tsezou A. Two Truncating MITF Variants Expand the Phenotypic Spectrum of COMMAD Syndrome. Clin Genet. 2026. doi:10.1111/cge.70260. PMID: 42843817.
72. Morell R, Spritz RA, Ho L, Pierpont J, Guo W, Friedman TB, et al. Apparent digenic inheritance of Waardenburg syndrome type 2 (WS2) and autosomal recessive ocular albinism (AROA). Hum Mol Genet. 1997;6(5):659-64. doi:10.1093/hmg/6.5.659. PMID: 9158138.
73. Yokoyama S, Woods SL, Boyle GM, Aoude LG, MacGregor S, Zismann V, et al. A novel recurrent mutation in MITF predisposes to familial and sporadic melanoma. Nature. 2011;480(7375):99-103. doi:10.1038/nature10630. PMID: 22080950.
74. Bertolotto C, Lesueur F, Giuliano S, Strub T, de Lichy M, Bille K, et al. A SUMOylation-defective MITF germline mutation predisposes to melanoma and renal carcinoma. Nature. 2011;480(7375):94-8. doi:10.1038/nature10539. PMID: 22012259.
75. National Center for Biotechnology Information. ClinVar; VCV000016633 — NM_001122659.3(EDNRB):c.828G>T (p.Trp276Cys) [Internet]. [accessed 2026-10-10]. Available at: https://www.ncbi.nlm.nih.gov/clinvar/variation/16633/
76. Xue C, Xu H, Huang X, Su M, He L, Zhang G, et al. Developmental arrest of astrocyte lineage in Snai2 deletion mice: implication for the intellectual disability in patients with Waardenburg syndrome. Transl Psychiatry. 2025;15(1):391. doi:10.1038/s41398-025-03616-9. PMID: 41073431. PMC12514258.
77. Clinical Genome Resource (ClinGen) Hearing Loss Gene Curation Expert Panel. Gene-disease validity curations: PAX3, MITF, SOX10, EDNRB, EDN3, SNAI2, KITLG [Internet]. Bethesda (MD): ClinGen; 2017–2024 [accessed 2026-10-10]. Available at: https://search.clinicalgenome.org/kb/gene-validity
78. Wildhardt G, Zirn B, Graul-Neumann LM, Wechtenbruch J, Suckfull M, Buske A, et al. Spectrum of novel mutations found in Waardenburg syndrome types 1 and 2: implications for molecular genetic diagnostics. BMJ Open. 2013;3:e001917. doi:10.1136/bmjopen-2012-001917. PMID: 23512835.
79. Liu D, Li A, Yu C, Ding X, Sun F, Qian X, et al. [Incomplete penetrance of SOX10 gene mutation in a family with Waardenburg syndrome type Ⅳ]. Lin Chuang Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2026;40(3):296-300. doi:10.13201/j.issn.2096-7993.2026.03.014. PMID: 41668496. PMC13368426.
80. National Center for Biotechnology Information. ClinVar; VCV000547774 — NM_006941.4(SOX10):c.127C>T (p.Arg43Ter) [Internet]. [accessed 2026-10-10]. Available at: https://www.ncbi.nlm.nih.gov/clinvar/variation/547774/
81. Ersson B, Gustafson E, Danielson J, Alafuzoff I. Pathology Seen in Myenteric Plexus in Two Subjects With Waardenburg Syndrome. Neurogastroenterol Motil. 2025;37(12):e70073. doi:10.1111/nmo.70073. PMID: 40364458. PMC12623266.
82. Bocángel MAP, Melo US, Alves LU, Pardono E, Lourenço NCV, Marcolino HVC, et al. Waardenburg syndrome: Novel mutations in a large Brazilian sample. Eur J Med Genet. 2018;61(6):348-354. doi:10.1016/j.ejmg.2018.01.012. PMID: 29407415.
83. National Center for Biotechnology Information. ClinVar; VCV004070994 — NM_207034.3(EDN3):c.472C>T (p.Arg158Cys) [Internet]. [accessed 2026-10-10]. Available at: https://www.ncbi.nlm.nih.gov/clinvar/variation/4070994/
84. Yang S, Dai P, Liu X, Kang D, Zhang X, Yang W, et al. Genetic and phenotypic heterogeneity in Chinese patients with Waardenburg syndrome type II. PLoS One. 2013;8:e77149. doi:10.1371/journal.pone.0077149. PMID: 24194866.
85. Pardono E, van Bever Y, van den Ende J, Havrenne PC, Iughetti P, Maestrelli SR, et al. Waardenburg syndrome: clinical differentiation between types I and II. Am J Med Genet A. 2003;117A(3):223-35. doi:10.1002/ajmg.a.10193. PMID: 12599185.
86. Tamayo ML, Gelvez N, Rodriguez M, Florez S, Varon C, Medina D, et al. Screening program for Waardenburg syndrome in Colombia: clinical definition and phenotypic variability. Am J Med Genet A. 2008;146A:1026-31. doi:10.1002/ajmg.a.32189. PMID: 18241065.
87. Newton VE. Clinical features of the Waardenburg syndromes. Adv Otorhinolaryngol. 2002;61:201-8. doi:10.1159/000066810. PMID: 12408085.
88. Li MM, Tayoun AA, DiStefano M, Pandya A, Rehm HL, Robin NH, et al. Clinical evaluation and etiologic diagnosis of hearing loss: A clinical practice resource of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2022;24(7):1392-1406. doi:10.1016/j.gim.2022.03.018. PMID: 35802133.
89. Mahoney K, Crocker K, Blair J, Carratu K, Araya G, Kaur M, et al. Revisiting the W-Index and Waardenburg Syndrome: A Retrospective Review of Waardenburg Syndrome Diagnoses at a Single Site Hearing Loss Clinic and the Sensitivity, Specificity, and Genotype-Phenotype Correlations of an Elevated W-Index. Am J Med Genet A. 2026;200(7):1551-1562. doi:10.1002/ajmg.a.70096. PMID: 41795619.
90. Madden C, Halsted MJ, Hopkin RJ, Choo DI, Benton C, Greinwald JH Jr. Temporal bone abnormalities associated with hearing loss in Waardenburg syndrome. Laryngoscope. 2003;113:2035-41. doi:10.1097/00005537-200311000-00034. PMID: 14603070.
91. Black FO, Pesznecker SC, Allen K, Gianna C. A vestibular phenotype for Waardenburg syndrome? Otol Neurotol. 2001;22:188-94. doi:10.1097/00129492-200103000-00012. PMID: 11300267.
92. Elmaleh-Bergès M, Baumann C, Noel-Petroff N, Sekkal A, Couloigner V, Devriendt K, et al. Spectrum of temporal bone abnormalities in patients with Waardenburg syndrome and SOX10 mutations. AJNR Am J Neuroradiol. 2013;34:1257-63. doi:10.3174/ajnr.A3367. PMID: 23237859.
93. Shin KH, Kim CH, Kim Y, Lee JK, Park HR, Han JH, et al. Cochlear duct length variability in prelingual sensorineural hearing loss: Emphasis on Waardenburg syndrome and implications for cochlear implantation. Eur Arch Otorhinolaryngol. 2026;283(5):3051-3060. doi:10.1007/s00405-026-10046-w. PMID: 41786931. PMC13152929.
94. Botros S, Bechara Ghobril A, Shields CL. Uveal Hypopigmentation Revealing Waardenburg Syndrome. JAMA Ophthalmol. 2026. doi:10.1001/jamaophthalmol.2026.3756. PMID: 42720929.
95. Ramtohul P, Hereth E, David T. Foveal Hypoplasia in Waardenburg Syndrome. Ophthalmology. 2026;133(7):e99. doi:10.1016/j.ophtha.2026.02.005. PMID: 41823913.
96. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Other cutaneous anomalies; p. 969.
97. Online Mendelian Inheritance in Man, OMIM®. Johns Hopkins University, Baltimore (MD). MIM #193500 — Waardenburg syndrome, type 1; WS1 [Internet]. [accessed 2026-10-10]. Available at: https://www.omim.org/entry/193500
98. Kujat A, Veith VP, Faber R, Froster UG. Prenatal diagnosis and genetic counseling in a case of spina bifida in a family with Waardenburg syndrome type I. Fetal Diagn Ther. 2007;22:155-8. doi:10.1159/000097117. PMID: 17139175.
99. Lemay P, Guyot MC, Tremblay É, Dionne-Laporte A, Spiegelman D, Henrion É, et al. Loss-of-function de novo mutations play an important role in severe human neural tube defects. J Med Genet. 2015;52:493-7. doi:10.1136/jmedgenet-2015-103027. PMID: 25805808.
100. Chen Q, Shi L, Chen Y, Cao X, Yang Y. Mild features of partial PAX3 deletion in patients with prenatal Waardenburg syndrome: a case report and literature review. Front Pediatr. 2025;13:1642132. doi:10.3389/fped.2025.1642132. PMID: 41141997. PMC12549294.
101. Banhos MEH, Barbosa MA, Ribeiro EM. Defeitos do fechamento do tubo neural. In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 160.
102. Online Mendelian Inheritance in Man, OMIM®. Johns Hopkins University, Baltimore (MD). MIM #609136 — Peripheral demyelinating neuropathy, central dysmyelination, Waardenburg syndrome, and Hirschsprung disease; PCWH [Internet]. [accessed 2026-10-10]. Available at: https://www.omim.org/entry/609136
103. Bogdanova-Mihaylova P, Alexander MD, Murphy RPJ, Murphy SM. Waardenburg syndrome: a rare cause of inherited neuropathy due to SOX10 mutation. J Peripher Nerv Syst. 2017;22(3):219-223. doi:10.1111/jns.12221. PMID: 28544110.
104. Yang Y, Zhang W, Zheng S, Jiang Y, Ye L, Sun S. SOX10 Mutation of Waardenburg Syndrome With Hypogonadism: A Report of 2 Cases. JCEM Case Rep. 2026;4(2):luaf339. doi:10.1210/jcemcr/luaf339. PMID: 41684881. PMC12891906.
105. Aboagye ET, Wonkam RP, de Kock C, Dandara C, Wonkam A. Genetics of Waardenburg Syndrome in Africa: A Systematic Review. Int J Mol Sci. 2025;27(1). doi:10.3390/ijms27010127. PMID: 41516007. PMC12785251.
106. Joint Committee on Infant Hearing. Year 2019 Position Statement: Principles and Guidelines for Early Hearing Detection and Intervention Programs. J Early Hear Detect Interv. 2019;4(2):1-44. doi:10.15142/fptk-b748. Available at: https://digitalcommons.usu.edu/jehdi/vol4/iss2/1/ [accessed 2026-10-10].
107. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Ocular hypertelorism; p. 258.
108. ClinicalTrials.gov. Search “Waardenburg” (condition), API v2 [Internet]. Bethesda (MD): National Library of Medicine; [accessed 2026-10-10]. Available at: https://clinicaltrials.gov/search?cond=Waardenburg (NCT02418936; NCT03655223)
109. Poon JYY, Yiu WC, Ho SKL, Cheng SSW, Luk HM, Lo IFM. Waardenburg Syndrome: Review of Genotype-Phenotype Relationships in 30 Patients in Hong Kong. Am J Med Genet A. 2025;197(12):e64193. doi:10.1002/ajmg.a.64193. PMID: 40702859.
110. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17:405-24. doi:10.1038/gim.2015.30. PMID: 25741868.
111. Zani A, Eaton S, Morini F, Puri P, Rintala R, van Heurn E, et al. European Paediatric Surgeons' Association Survey on the Management of Hirschsprung Disease. Eur J Pediatr Surg. 2017;27(1):96-101. doi:10.1055/s-0036-1593991. PMID: 27898990.
112. Jan IA, Stroedter L, Haq AU, Din ZU. Association of Shah-Waardenburg syndrome: a review of 6 cases. J Pediatr Surg. 2008;43:744-7. doi:10.1016/j.jpedsurg.2007.11.023. PMID: 18405726.
113. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 44, Disorders of cellular trafficking, 44.1.3.
114. National Organization for Rare Disorders. Waardenburg Syndrome [Internet]. Danbury (CT): NORD; updated 2026-02-26 [accessed 2026-10-10]. Available at: https://rarediseases.org/rare-diseases/waardenburg-syndrome/
115. Amirsalari S, Ajallouyean M, Saburi A, Haddadi Fard A, Abed M, Ghazavi Y. Cochlear implantation outcomes in children with Waardenburg syndrome. Eur Arch Otorhinolaryngol. 2012;269:2179-83. doi:10.1007/s00405-011-1877-3. PMID: 22159916.
116. de Sousa Andrade SM, Monteiro AR, Martins JH, Alves MC, Santos Silva LF, Quadros JM, et al. Cochlear implant rehabilitation outcomes in Waardenburg syndrome children. Int J Pediatr Otorhinolaryngol. 2012;76:1375-8. doi:10.1016/j.ijporl.2012.06.010. PMID: 22784507.
117. Koyama H, Kashio A, Sakata A, Tsutsumiuchi K, Matsumoto Y, Karino S, et al. The hearing outcomes of cochlear implantation in Waardenburg syndrome. Biomed Res Int. 2016;2016:2854736. doi:10.1155/2016/2854736. PMID: 27376080.
118. Fan W, Ni K, Chen F, Li X. Hearing characteristics and cochlear implant effects in children with Waardenburg syndrome: a case series. Transl Pediatr. 2022;11:1234-41. doi:10.21037/tp-22-271. PMID: 35958009.
119. Vanstrum EB, Castellanos CX, Ziltzer RS, Ulloa R, Moen R, Choi JS, et al. Cochlear implantation in Waardenburg syndrome: Systematic review and meta-analysis. Int J Pediatr Otorhinolaryngol. 2023;175:111738. doi:10.1016/j.ijporl.2023.111738. PMID: 37847940.
120. Arai S, Tomioka R, Shirai K, Nishiyama N, Miyake K, Yano T, et al. Long-Term Hearing, Language, and Educational Outcomes After Cochlear Implantation in Children With Waardenburg Syndrome. Cureus. 2025;17(11):e97615. doi:10.7759/cureus.97615. PMID: 41450428. PMC12728366.
121. Rossi D, Irvine W, Granström AL, Mutanen A, Bakx R, Bjørnland K, et al. Updated European Reference Network for rare Inherited and Congenital Digestive and Gastrointestinal Anomalies guidelines for the management of rectosigmoid Hirschsprung's disease 2025. J Pediatr Gastroenterol Nutr. 2026;83(1):185-207. doi:10.1002/jpn3.70442. PMID: 42057764.
122. Fleming A, Copp AJ. Embryonic folate metabolism and mouse neural tube defects. Science. 1998;280:2107-9. doi:10.1126/science.280.5372.2107. PMID: 9641914.
123. Gironi LC, Zottarelli F, Esposto E, Cammarata E, Giorgione G, Mellone S, et al. Dermatologic surveillance in healthy carriers of CDKN2A and p.E318K MITF germline variants from melanoma-prone families: a 14 years hospital-based experience. Melanoma Res. 2025;35(6):424-428. doi:10.1097/CMR.0000000000001054. PMID: 40623218.
124. US Preventive Services Task Force; Barry MJ, Nicholson WK, Silverstein M, Chelmow D, Coker TR, Davis EM, et al. Folic Acid Supplementation to Prevent Neural Tube Defects: US Preventive Services Task Force Reaffirmation Recommendation Statement. JAMA. 2023;330(5):454-459. doi:10.1001/jama.2023.12876. PMID: 37526713.
125. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 500–504 (Neural tube defects).
126. Kapur S, Karam S. Germ-line mosaicism in Waardenburg syndrome. Clin Genet. 1991;39:194-8. doi:10.1111/j.1399-0004.1991.tb03011.x. PMID: 2036740.
127. Chen K, Zhan Y, Wu X, Zong L, Jiang H. Germinal mosaicism of PAX3 mutation caused Waardenburg syndrome type I. Int J Pediatr Otorhinolaryngol. 2018;104:200-4. doi:10.1016/j.ijporl.2017.11.011. PMID: 29287868.
128. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Genetic imbalance: single gene disorders; p. 941.
129. Huang SJ, Amendola LM, Sternen DL. Variation among DNA banking consent forms: points for clinicians to bank on. J Community Genet. 2022;13:389-97. doi:10.1007/s12687-022-00601-3. PMID: 35834113.
130. Contact. Waardenburg syndrome [Internet]. London: Contact a Family; [accessed 2026-10-10]. Available at: https://contact.org.uk/conditions/waardenburg-syndrome/
131. Guhan SM, Artomov M, McCormick S, Njauw C, Stratigos AJ, Shannon K, et al. Cancer risks associated with the germline MITF(E318K) variant. Sci Rep. 2020;10(1):17051. doi:10.1038/s41598-020-74237-z. PMID: 33051548.