Kabuki Syndrome

Literature and regulatory information updated: 6 October 2026.

Kabuki syndrome type 1 (KMT2D, autosomal dominant) and type 2 (KDM6A, X-linked); historical synonyms: Kabuki make-up syndrome, Niikawa-Kuroki syndrome

0. Initial Reference Table

1. Introduction

Kabuki syndrome (KS) is a multiple congenital anomaly and neurodevelopmental syndrome that Orphanet defines through five groups of major manifestations: intellectual disability, typically mild to moderate; visceral malformations, particularly congenital heart defects; persistent fetal fingertip pads; postnatal short stature and skeletal anomalies such as brachymesophalangy, fifth-finger brachydactyly, spinal anomalies and clinodactyly; and a characteristic facies, with broad arched eyebrows, long palpebral fissures, lower eyelid eversion, large, prominent or cupped ears, a depressed nasal tip and short columella [1]. Molecularly, it is now understood as a disorder of the epigenetic machinery, or chromatinopathy, caused by loss of function of two histone-modifying enzymes that collaborate in chromatin activation: the histone methyltransferase KMT2D in the autosomal dominant form, and the demethylase KDM6A in the X-linked form [2].

The history of the entity began in Japan in 1981, with two independent descriptions published in the same issue of the Journal of Pediatrics. Norio Niikawa and colleagues described five unrelated Japanese children with moderate to severe mental retardation, progressive postnatal-onset dwarfism, long palpebral fissures with eversion of the lateral third of the lower eyelid, arched eyebrows, a broad depressed nasal tip, large prominent ears, short fifth fingers, abnormal dermatoglyphics with absence of digital triradius c or d, and frequent otitis media in childhood; they found neither familial recurrence nor consanguinity and left the aetiology open [18]. Yoshikazu Kuroki and colleagues described another five patients with moderate mental retardation, postnatal dwarfism, susceptibility to childhood infections, long palpebral fissures, abnormal arched eyebrows, long eyelashes, large ears, a short nasal septum, cleft palate, short fingers and vertebral anomalies, and, combining their cases with Niikawa's, proposed a new nosological entity; they considered an environmental cause but did not exclude autosomal dominant or X-linked inheritance [19]. The name derives from the resemblance between patients' faces and the make-up of actors in traditional Japanese Kabuki theatre [20].

In 1988, a collaborative study involving 33 institutions brought together 62 patients, 58 Japanese, and defined the five cardinal manifestations that still underpin clinical recognition: peculiar facies (100%), skeletal anomalies (92%), dermatoglyphic abnormalities including fingertip pads (93%), mild to moderate mental retardation (92%) and postnatal growth deficiency (83%); in that study, all cases were sporadic, the sex ratio was balanced, prevalence was estimated at 1:32,000 Japanese newborns, and the hypothesis most compatible with the data was an autosomal dominant disorder in which each patient represented a new mutation [8]. Three of the 62 patients had a Y chromosome abnormality with a shared breakpoint at Yp11.2, leading the authors to consider pseudoautosomal inheritance, a hypothesis not confirmed by subsequent gene identification [8]. In the following decades, the syndrome was described in non-Japanese populations, and Smith records more than 400 published patients [20].

The molecular aetiology was clarified in 2010, in one of the earliest examples of identifying a Mendelian disease gene through exome sequencing: in ten unrelated probands, genotypic and phenotypic stratification pointed to MLL2 (now KMT2D), with nonsense or frameshift variants in seven; Sanger sequencing found variants in two of the remaining three and in 26 of 43 additional cases, and, where parental DNA was available, the variant was de novo in 12 families and transmitted in two, consistent with the phenotype [21]. In 2012, Lederer and colleagues described three patients with partial or complete de novo KDM6A deletions, a histone demethylase interacting with KMT2D, and observed preferential inactivation of the X carrying the deletion, despite the gene escaping X-inactivation [22]; in 2013, Miyake and colleagues identified the first KDM6A point variants in three of 32 patients without a KMT2D variant [23]. In 2019, an international expert group, drawing on 70 publications with at least one molecularly confirmed patient, proposed consensus diagnostic criteria allowing a definitive diagnosis with or without molecular confirmation, at a time when therapies targeting the epigenetic mechanism were anticipated [11]. Discovery of a specific DNA methylation signature in peripheral blood added a functional tool for diagnosis and variant reclassification from 2017 onwards [2,24].

The clinical importance of KS for the geneticist stems from four aspects: its relative frequency among malformation syndromes with intellectual disability; variation in facies over the lifespan and between genotypes, making neonatal recognition difficult; the presence of treatable complications whose omission worsens prognosis, particularly congenital hyperinsulinism, left-sided obstructive heart defects, humoral immunodeficiency and hearing loss; and the existence of differential diagnoses and allelic diseases with different mechanisms, requiring careful variant interpretation [2,25,26].

2. Molecular Genetics and Pathophysiology

2.1 Genetic architecture and proportion of cases

KS is genetically heterogeneous, but the heterogeneity is limited. GeneReviews attributes about 75% of cases with a diagnosis of KS to pathogenic variants in KMT2D and about 3–5% to variants in KDM6A; in approximately 30% of individuals with a clinical diagnosis the genetic cause remains unidentified, which keeps open the possibility of genes not yet described [2]. Published proportions vary with the denominator: Smith reports about 60% of cases explained by KMT2D and 10–20% of diagnosed patients with KDM6A alterations; the Oxford Desk Reference gives KMT2D variants in 50–75% of cases with a secure clinical diagnosis; the 2019 consensus reports 70% with a variant in KMT2D or KDM6A; and, among patients with molecular confirmation, more than 80% have variants in KMT2D and 5–11% in KDM6A [11,20,25,27]. These differences are not contradictory: they reflect the mix of patients with typical and atypical phenotypes in the series, the rigour of the clinical diagnosis and the detection methods used. Banka and colleagues, in a cohort of 116 patients, observed that almost all patients with the typical facies had a pathogenic variant in KMT2D and concluded that the genetic heterogeneity of KS is lower than had been assumed [28]. Patients with a clinical diagnosis of KS and negative testing of both genes are described in the literature as having a “Kabuki-like” phenotype; the causes identified in these patients are described in sections 6.2 to 6.4 [14].

2.2 KMT2D: structure, function and variant spectrum

The KMT2D gene (12q13.12; synonyms MLL2, ALR, MLL4) encodes a histone lysine methyltransferase transferring methyl groups from S-adenosyl-L-methionine to lysine 4 of histone H3 (H3K4) [2,4]. Through its SET domain, the enzyme catalyses conversion of unmethylated H3K4 into mono-, di- and trimethylated forms, predominantly depositing the H3K4me1 mark characteristic of active enhancers [2,4]. KMT2D forms part of the ASCOM (activating signal cointegrator-2-containing complex) complex, within the COMPASS family, which removes repressive epigenetic marks and deposits activating methylation marks, recruiting RNA polymerase II and producing an active chromatin state [2]. Most target genes and regulated pathways remain to be characterised [2]. The KS1 mechanism is loss of function through haploinsufficiency: ClinGen classifies KMT2D haploinsufficiency evidence as sufficient, does not recognise triplosensitivity, and classified the KS1 gene–disease relationship as definitive [6].

Most KS1 pathogenic variants introduce a premature termination codon (nonsense, frameshift or splicing variants) and are distributed throughout the gene [28,29]. Missense variants account for 15–20% of cases according to Baldridge and colleagues, or 15–30% according to a 2026 mouse-model study [30,31]; Banka and colleagues observed that pathogenic missense variants were frequently located in exon 48 [28]. Intragenic or whole-gene deletions, somatic mosaicism (including mildly affected parents) and, in 2025, the first insertion of an AluY element into KMT2D have been described, the latter identified by “reverse genotyping” in an undiagnosed patient in whom the methylation signature was positive [2,32,33]. In a mouse model carrying a patient missense variant at residue Arg5179, NM_003482.4:c.15536G>A p.(Arg5179His) [5] (corresponding to R5230H in the mouse, historical notation), the substitution did not alter stability or enzymatic activity in vitro, but reproduced the growth deficit, craniofacial anomalies and IgA deficiency, without changes in neurological function, and revealed perinatal lethality and unilateral renal agenesis, suggesting distinct mechanisms for some missense variants [31].

2.3 KDM6A: structure, function, X-inactivation and variant spectrum

The KDM6A gene (Xp11.3; synonym UTX) encodes a demethylase with a Jumonji C (JmjC) domain removing di- and trimethylated marks from lysine 27 of histone H3 (H3K27), deposited by Polycomb complexes; its activity accompanies H3K4 methylation, controls PRC1 complex recruitment and histone H2A monoubiquitination, regulates HOX genes and remodels chromatin for T-box factor-dependent expression, partly independently of demethylase activity [2,4]. At its amino terminus, the protein contains eight tetratricopeptide repeats (TPR) of incompletely understood function [3]. KDM6A escapes X chromosome inactivation; nevertheless, Lederer and colleagues observed skewed inactivation of the X carrying the deletion in female patients [2,22]. In the largest published series (80 patients, 34 male and 46 female, with 61 distinct pathogenic variants), 50 variants were truncating and 11 missense; among truncating variants, 30% were nonsense, 28% affected canonical splicing sites, 24% caused frameshifts, 16% were larger deletions and one resulted from a gene-disrupting translocation; missense variants clustered in TPR domains 2, 3 and 7 and JmjC; truncating variants were significantly more frequently de novo (87.5% versus 57.1% of protein-altering non-truncating variants; p = 0.02) [3]. Maternally inherited variants were identified in 13 patients and, for the first time, a variant inherited from an equally affected father (p.Ala149Thr, using NM_021140.3 numbering) [3]. GeneReviews adds that in available series, variants at the 3' end of the gene are more frequent than at the 5' end, and splicing variants are the most common type of single-nucleotide variant [2]. Around 20% of pathogenic KDM6A variants are intragenic deletions or duplications, with direct implications for test selection [2].

2.4 Variant nomenclature and reference transcripts

The KS literature uses NM_003482.3 or NM_003482.4 for KMT2D and almost always NM_021140.3 for KDM6A [3,25]. For KMT2D, the MANE Select (Matched Annotation from NCBI and EMBL-EBI) transcript is NM_003482.4, and coding positions of variants described on NM_003482.3 remain unchanged in the current version. For KDM6A, however, the MANE Select transcript is NM_001291415.2, with a longer coding sequence, so numbering diverges downstream of codon 362: the nonsense variant described as c.3717G>A p.(Trp1239Ter) on NM_021140.3 corresponds to c.3873G>A p.(Trp1291Ter) on NM_001291415.2 [5]. A genetic report should therefore always specify the versioned transcript; comparison with published variants must use genomic coordinates or validated conversion, never manual calculation. Table 2.1 presents examples normalised using VariantValidator.

Table 2.1 — Examples of published variants normalised using VariantValidator (GRCh38)

2.5 Pathophysiology: from chromatin imbalance to organs

KMT2D and KDM6A act on the same side of the balance between open and closed chromatin: one deposits the activating H3K4 mark and the other removes the repressive H3K27me3 mark, explaining why loss of function of either produces overlapping phenotypes [2,4]. In a mouse model heterozygous for Kmt2d (Kmt2d+/βGeo), Bjornsson and colleagues demonstrated reduced histone H4 acetylation and H3K4 trimethylation in fibroblasts, H3K4me3 deficiency in the dentate gyrus granule cell layer, reduced adult neurogenesis and hippocampal memory deficits; these alterations improved with the histone deacetylase inhibitor AR-42 administered after birth, suggesting that part of the intellectual disability depends on a reversible deficit in postnatal neurogenesis [36]. In the same model, a ketogenic diet, through increased beta-hydroxybutyrate (an endogenous deacetylase inhibitor), modulated H3 acetylation and H3K4me3 and corrected neurogenesis and memory deficits [37]; and the KDM1A (LSD1) demethylase inhibitor TAK-418, which prevents removal of the H3K4 marks deposited by KMT2D, increased newly formed doublecortin-positive hippocampal cells in a dose-dependent manner, corrected histone alterations and immediate-early gene expression changes such as Fos and Fosb, and normalised hippocampal memory after two weeks [38].

In the immune system, KMT2D haploinsufficiency disrupts terminal B-lymphocyte differentiation: in a cohort of 13 patients, hypogammaglobulinaemia was present in all but one carrier of a KMT2D variant, IgA deficiency affected 90%, and memory and class-switched memory B cells were reduced [39]. In the mouse model, Kmt2d deficiency reduced IgA-secreting cells, Peyer's patch number and size, and expression of Itgb7, which mediates lymphocyte intestinal homing, a finding confirmed in human samples [40]. The mechanism of hyperinsulinism remains unclear; epigenetic alterations during pancreatic differentiation and abnormal regulation of insulin secretion have been proposed, and KDM6A inhibition increases insulin release in mouse pancreatic islets [3,25].

Chromatin disruption leaves a measurable mark in peripheral blood: the DNA methylation signature, or episignature, which Aref-Eshghi and colleagues described as an "epigenetic echo" of the disease, with alterations at 1504 CpG sites and 24 genomic regions, including several HOX genes and MYO1F [24]. Butcher and colleagues showed that KS1 and CHARGE syndrome (CHD7) signatures are distinct, although they share targets such as HOXA5, which may explain some clinical overlap between the two syndromes [41]. The diagnostic application of this signature is addressed in section 5.

2.6 Genotype–phenotype correlations

Patients with a KMT2D variant more often have the typical facies, which may partly reflect the inclusion of misclassified patients in the variant-negative groups; they also have more renal anomalies, feeding difficulties, premature thelarche, joint dislocations and palatal anomalies than patients without a KMT2D variant [2,28]. In the cohort of 86 clinically diagnosed patients in whom Makrythanasis and colleagues prospectively applied their phenotypic score, ten features were more frequent in KMT2D variant carriers than in variant-negative patients: blue sclerae, arched eyebrows, broad nasal root, depressed nasal tip, large dysplastic ears, thin upper-lip vermilion with a thick lower lip, joint hyperlaxity, short stature, frequent infections and intellectual disability [11]. The characteristic dental anomalies (absence of the upper lateral incisors, lower incisors and/or second premolars, and flat-blade screwdriver-shaped upper incisors) may also be more frequent in KMT2D variant carriers [11]. Whole-gene deletions and truncating variants in the first half of KMT2D may be associated with more severe intellectual disability, and missense variants in the terminal regions with a greater risk of autoimmune disease [2]. In a French series of 18 patients, all six with neonatal hypoglycaemia had nonsense or frameshift variants in the carboxy-terminal portion of the protein [42]. In the next-generation episignature published in 2026, missense variants in exon 48 and in the amino-terminal PHD-type zinc fingers were predominantly classified as pathogenic for KS1, whereas those in the central region (exons 31–39) were more often classified as benign for KS1, consistent with the association of these variants with another syndrome [43].

In KS2, hyperinsulinaemic hypoglycaemia, hypertrichosis, long great toes and large central incisors are proportionally more frequent [2,7]. In absolute numbers, more patients have hyperinsulinism and a KMT2D variant, because this gene explains many more cases [2]. Sex strongly modulates the phenotype: in the Faundes series, boys were born earlier (median 36 versus 39 weeks), had shorter final height (median −2.18 versus −1.2 standard deviations), walked (50% versus 95.5%) and developed language (50% versus 90.9%) less frequently, and more frequently had severe intellectual disability (77.8% versus 33.3%) [3]. This difference occurs despite KDM6A escaping X-inactivation and has no established explanation in the sources consulted [2,3]; the Y chromosome paralogue, KDM6C (UTY), has been studied in boys with KS and is listed by GeneReviews among candidate genes or genes with overlapping phenotypes [2,44].

2.7 Allelic diseases and related entities

KMT2D missense variants in exons 38 and 39 cause a syndrome distinct from KS. GeneReviews places them in a region of about 56 amino acids between Leu3525 and Lys3583; Cuvertino and colleagues describe a highly conserved 54-amino-acid region between Val3527 and Lys3583, and Baldridge and colleagues a 40-amino-acid region immediately amino-terminal to a coiled-coil domain; the differences reflect the boundaries adopted in each series [2,26,30]. The picture includes choanal atresia, athelia or hypoplastic nipples, branchial sinus anomalies, cervical pits, lacrimal duct anomalies, hearing loss with external and inner ear malformations, thyroid abnormalities, hypoparathyroidism, interstitial lung disease, extreme short stature and delayed or absent puberty, without the typical KS facies; in the Cuvertino series no patient had intellectual disability, whereas GeneReviews reports variable developmental delay or intellectual disability [2,26,30]. The frequency of manifestations, automated facial analysis and DNA methylation pattern differ significantly from KS1, and the proposed mechanism is dominant negative or gain of function, in contrast with the haploinsufficiency of KS1 [2,26]. This entity has its own OMIM number (620186) and is designated by the acronym BCAHH (branchial arch abnormalities, choanal atresia, athelia, hearing loss and hypothyroidism); ERN ITHACA issued a call for collaboration to delineate it [4,45]. The practical point is that these patients may formally meet the KS consensus criteria (hypotonia or developmental delay plus a pathogenic variant in KMT2D), even though the phenotype is different [2].

Alobar holoprosencephaly has also been described in two fetuses with de novo KMT2D variants (one nonsense and one missense), without KS traits; this is an observation of isolated cases without ClinGen curation or an OMIM entry [4,46]. Somatic KMT2D variants arise in various tumours (central nervous system, gastrointestinal and haematological), and KDM6A variants predominantly in urothelial carcinoma, T-cell acute lymphoblastic leukaemia and breast cancer; no germline KDM6A phenotypes other than KS are known [2]. Bögershausen and colleagues discussed a possible role of KDM6A in patients with Turner syndrome and Kabuki traits, and Saudubray cites the association between congenital hyperinsulinism in Turner syndrome and KDM6A haploinsufficiency [44,47].

2.8 Penetrance and expressivity

Penetrance of pathogenic KMT2D variants appears complete; for KDM6A, data are insufficient. Variable expressivity may lead to underdiagnosis of mildly affected individuals, including parents with somatic mosaicism [2].

3. Clinical Spectrum and Manifestations by System

3.1 Overview

The GeneReviews clinical description summarises more than 400 molecularly confirmed individuals and organises the phenotype into three layers: the five cardinal manifestations defined by Niikawa (typical facies, skeletal anomalies, dermatoglyphic abnormalities, mild to moderate intellectual disability and postnatal growth deficiency), structural anomalies (ophthalmological, ear pits, cleft lip and/or palate, dental anomalies, heart defects, gastrointestinal and genitourinary anomalies), and functional differences (hearing loss, feeding difficulties, endocrine abnormalities including hyperinsulinism and isolated premature thelarche, susceptibility to infections and autoimmune disease, seizures) [2]. Frequencies vary with the period, inclusion criteria and cohort origin; Table 3.1 brings together estimates from the main sources, with their denominators, and should not be read as population prevalence.

Table 3.1 — Frequency of main manifestations by source

The systematic review underpinning the 2019 consensus criteria compiled 449 molecularly confirmed patients described in 70 publications: 399 with a pathogenic variant in KMT2D (172 female, 167 male and 60 with sex not stated) and 50 with a pathogenic variant in KDM6A (30 female and 20 male) [11]. Table 3.1b reproduces the number of patients in whom each feature was described. Because not every feature was sought or reported in every patient, the values are numerators without a uniform denominator and cannot be converted into frequencies; the authors note, for example, that malpositioned kidneys are underestimated, because many publications reported only “renal anomalies” without specifying the type [11].

Table 3.1b — Number of molecularly confirmed patients in whom each feature was described (review of 70 publications; 2019 consensus, Figure 2)

Source: [11]. The numbers indicate patients in whom the feature was reported, not patients assessed for that feature.

3.2 Facies and changes with age

The typical facies comprises long palpebral fissures (measurement two or more standard deviations above the mean for age) with eversion of the lateral third of the lower eyelid, arched and broad eyebrows with a sparse or notched lateral third, a short columella with a depressed nasal tip and large, prominent or cupped ears [2]. Smith adds ptosis, bluish sclerae, strabismus, epicanthic folds, preauricular pits, cleft palate, dental anomalies and an open mouth with a tented upper lip, which gives a myopathic appearance [20]. Figures 1 and 2 of Smith (pp. 140–1) show these features in an 18-month-old boy and in two children aged 3 and 4 years [20]. Although many features are present in the newborn, they become more evident with age, and the Oxford Desk Reference stresses that the diagnosis is usually made only when the facies becomes characteristic, at around 12 months [20,52]. The 2019 international consensus places between 3 and 12 years the age at which the facies is easiest to recognise and most classic; in infancy it may be difficult to identify and, in adolescence and adulthood, the lower-eyelid eversion may disappear, which is why the clinical criterion accepts features documented at any point in life [11]. To measure the palpebral fissure, the examiner ideally sits at the patient's eye level; the patient, if cooperative, keeps the head in a neutral position and looks at the ceiling; the distance between the inner and outer canthus is measured with a transparent ruler and the result is compared with age-specific charts (Figure 3 of the consensus article illustrates the technique: https://doi.org/10.1136/jmedgenet-2018-105625) [11]. In adults, according to the 2010 European guideline, truncal obesity appears, the eyebrows no longer look notched and the foot features become more marked, with overlapping toes, a long hallux and relatively short fourth and fifth toes [12]. In KS2 the facies is more variable: notched eyebrows in only some patients, often thick, less prominent and fleshier ears, a prominent lower lip and a prominent forehead; Figure 2 of Faundes and colleagues illustrates this variability by sex and variant type (https://pmc.ncbi.nlm.nih.gov/articles/PMC8257478/figure/Fig2/) [3].

3.3 Growth, feeding and body composition

Birth parameters are usually normal; in infancy, feeding difficulties (around 70%) relate to hypotonia, poor oromotor coordination and dysphagia, may require a nasogastric tube or gastrostomy, and gastro-oesophageal reflux is frequent [2]. Without treatment, postnatal growth deficiency is evident at 12 months and worsens through absence of a typical pubertal growth spurt; in a series of 39 patients with a KMT2D variant without hormonal treatment, all had postnatal growth retardation and a reduced pubertal growth spurt [2,53]. Using 95 French patients with KS1, Ruault and colleagues constructed syndrome-specific height, weight, body mass index and head circumference charts; mean height was 2 standard deviations (boys) and 1.8 standard deviations (girls) below parental target height [54]. Microcephaly may or may not accompany short stature [2]. More than half of adolescents and adults develop obesity, which worsens problems such as recurrent patellar dislocation [2]; in the Australian and New Zealand series, 8 of 14 patients older than 5 years were overweight or obese, and 6 of these had experienced failure to thrive in childhood, an evolutionary profile the authors considered characteristic [9]. The first growth charts from a Chinese cohort (104 children, 433 height measurements) were published in 2025 [49].

3.4 Neurodevelopment, cognition and behaviour

Hypotonia is present in most children and tends to improve with age [2]. Intellectual disability is usually mild to moderate: Smith reports a mean developmental quotient of 52 in infants and children and a mean IQ of 57 in older patients, with an IQ of 80 or above in 12% and normal intelligence possible; GeneReviews reports mean IQs between the high 50s and the high 60s in KMT2D variant carriers and mentions rare patients with an IQ above 70 [2,20]. Most patients speak and walk [2]. In the 2019 consensus review, an Italian cohort of 17 patients with standardised neuropsychiatric assessment (9 with a KMT2D variant and 8 with a clinical diagnosis) showed no differences in neuropsychological profile between the two groups, and 11 of the 17 (66%) had an IQ below 80, similar to the 84% with an IQ below 80 in the Japanese series of Matsumoto and Niikawa; not all patients have intellectual disability, the severe form is uncommon and weakness of motor skills, attributed to the combination of joint hyperlaxity and hypotonia, is marked [11]. On neuropsychological testing, results are better for vocabulary comprehension and working memory and worse for non-verbal reasoning and processing speed; all linguistic subdomains may be affected and dysarthria (reduced rate and stress, distorted pitch, harsh voice, hypernasality, imprecise consonants) has been described [2]. Visuomotor and visuoperceptual difficulties may compound learning problems, so audio-verbal learning is suggested [2]. Patients are usually described as friendly and sociable; attention deficit hyperactivity disorder occurs in a subgroup, and anxiety, self-injury and sleep disturbance are rare [2]. In a sample of 35 patients assessed with social responsiveness scales, about one third had a prior diagnosis of autism spectrum disorder (67% in KS2 versus 23% in KS1), 72% were at borderline or clinical levels for peer problems and only 3% had atypical results for prosocial behaviour [55]. In KS2, boys have severe intellectual disability more often than girls, some of whom have normal development [3,7].

3.5 Nervous system

Seizures occur in 10–39% of patients, including infantile spasms, and are generally controlled with conventional antiseizure medicines [2]. Major brain malformations are rare; cerebellar and brainstem atrophy, Dandy-Walker malformation, delayed myelination and mild ventriculomegaly have been described, and, before the molecular era, symptomatic Chiari I malformation [2]. Smith also lists microcephaly, craniosynostosis, polymicrogyria, subarachnoid cyst and hydrocephalus due to aqueductal stenosis as occasional findings [20].

3.6 Heart and vessels

Around 70% of patients with molecularly confirmed KS1 have congenital heart disease, predominantly left-sided obstructive lesions, particularly aortic coarctation, sometimes in associations resembling Shone complex [2,10]. In the Digilio and colleagues series (19 of 27 patients with a KMT2D variant), the most frequent types were aortic coarctation (21%), bicuspid aortic valve (21%), subaortic perimembranous ventricular septal defect (16%), ostium secundum atrial septal defect (16%) and conotruncal heart defects (16%), with aortic dilatation with mitral abnormality and hypoplastic left heart syndrome also described [10]. Smith adds mitral valve prolapse, pulmonary, aortic and mitral valvular stenoses, tetralogy of Fallot, single ventricle with a common atrium, double-outlet right ventricle and transposition of the great arteries, and GeneReviews reports occasional hypertrophic cardiomyopathy and aortic root dilatation [2,20]. Frequency varies greatly between cohorts (27% in a Chinese cohort, with anomalous pulmonary venous connection in three patients, 49% in a North American cohort of 37 patients) [49,50].

3.7 Kidney and urinary tract

Renal and urinary tract anomalies occur in more than 25% of patients: positional and ascent anomalies (single fused kidney, crossed fused renal ectopia), renal dysplasia, hydronephrosis (the most common urinary finding), ureteropelvic junction obstruction and collecting system duplication; hypospadias and cryptorchidism occur in boys [2]. A Korean cohort of 65 patients (59 with a KMT2D variant) followed to a median age of 9.4 years showed congenital kidney and urinary tract anomalies in 33.9%, nephrolithiasis and/or nephrocalcinosis in 11.3%, and progression to chronic kidney disease in 33.9% (19 of 56); 25% and 50% of patients reached stage G2 at median ages of 5.8 and 24.6 years, and younger age at diagnosis and bilateral renal anomalies predicted progression, with congenital anomalies and cardiorenal syndrome the main causes [51].

3.8 Eyes, ears, mouth and teeth

Ocular findings affect more than one third of patients: bluish sclerae, strabismus, ptosis, coloboma, Marcus Gunn phenomenon and corneal anomalies such as Peters anomaly; optic nerve hypoplasia, colobomatous microphthalmia and anophthalmia occur rarely [2]. Eyelid eversion causes epiphora, and nocturnal lagophthalmos, which is frequent, can cause corneal abrasions and scarring [2]. Chronic otitis media is an important cause of morbidity and conductive hearing loss; sensorineural hearing loss is less frequent, may be progressive and is associated with inner ear malformations (Mondini dysplasia, vestibular enlargement, aplasia of the cochlea and semicircular canals, enlarged aqueduct) [2]; these malformations enter the differential diagnosis of genetic deafness with enlarged vestibular aqueduct. About one third have cleft lip and/or palate, submucous cleft may be underdiagnosed, almost three quarters have a high palate and some have lower-lip pits [2]. Hypodontia is the most common dental anomaly (absent upper lateral incisors and lower incisors, absent second premolars), together with abnormally shaped teeth (screwdriver-shaped upper incisors), small or widely spaced teeth and malocclusion [2].

3.9 Gastrointestinal and respiratory systems

Structural gastrointestinal anomalies are uncommon but include anorectal malformations (imperforate anus, anovestibular fistula, anterior anus), congenital diaphragmatic hernia and diaphragmatic eventration, cholestasis of various causes, and chronic diarrhoea due to malabsorption or coeliac disease [2]. Smith also reports intestinal malrotation, biliary atresia and bronchial tree stenosis, and the 2010 European guideline recommends considering extrahepatic biliary atresia with prolonged jaundice [12,20]. The Oxford Desk Reference describes diaphragmatic hernia or eventration as a relatively rare presentation [52]. Laryngeal anomalies may complicate anaesthesia [2].

3.10 Musculoskeletal system and skin

Joint hypermobility affects 50–75% of patients and is associated with hip, patellar and shoulder dislocations, improving with age [2]. Scoliosis and kyphosis may be associated with vertebral anomalies (hemivertebrae, butterfly vertebrae, sagittal clefts) [2]. Persistent fetal fingertip pads, fifth-finger brachydactyly, brachymesophalangy and fifth-finger clinodactyly are diagnostic clues; absence of digital triradii c and/or d and increased digital and hypothenar loops are rarely assessed in current practice [2]. Pilomatricomas, benign hair follicle tumours, are rare and resolve with excision [2]. The European guideline reports frequent flat feet, small or dystrophic nails and hirsutism in around one fifth of patients [12].

3.11 Endocrine system and hyperinsulinism

Premature thelarche is the most frequently described endocrine abnormality (16–41%), does not represent precocious puberty and tends to regress [2]. Adrenal insufficiency, combined pituitary deficiency, diabetes insipidus, growth hormone deficiency, hypothyroidism, primary ovarian dysfunction and true precocious puberty have been described in a small subgroup [2]. Of 18 children with KS1 undergoing stimulation tests, 5 (27.8%) met biochemical criteria for growth hormone deficiency, without correlation with IGF-1, leading the authors to question the utility of these tests in this syndrome [56].

Congenital hyperinsulinism (HI) deserves attention because it is treatable and unrecognised hypoglycaemia may cause irreversible neurological injury that worsens intellectual disability [2]. Around 1% of newborns with HI are estimated to have KS: in a series of 9 patients with KS and persistent HI, with KMT2D (5) and KDM6A (4) variants, reanalysis of 100 infants with HI without a genetic diagnosis revealed KS in one [57]. The largest cohort, from the HI centre at Children's Hospital of Philadelphia (1998–2023), included 33 children with KS and HI: hypoglycaemia was recognised on the first day of life in 25 (76%), but HI was diagnosed only at a median age of 1.8 months (interquartile range 0.6–6.1), in 6 children after 1 year of age and, in one case, at 28.3 months; in 61%, HI was diagnosed before KS, whose diagnosis occurred at a median age of 5 months [25]. Of 29 molecularly confirmed patients, 24 (83%) had a KMT2D variant and 5 (17%) a KDM6A variant, a proportion not significantly different from that expected (17% versus 11%; p = 0.31), unlike earlier series in which KDM6A explained 39–50% of HI cases with KS [25]. Diazoxide was tried in 25 children, with a response in 23 (92%), and treatment was discontinued in 15 (46%) at a median age of 2.8 years (range 0.6–10.5) [25]. In KS2, the Faundes series describes neonatal hypoglycaemia in 56.4% and HI in 27.7%, and the Oxford Desk Reference highlights the greater propensity of patients with KDM6A to present with hypoglycaemia [3,58]. Saudubray includes KS, alongside Beckwith-Wiedemann and Turner syndromes, among the most frequent causes of syndromic HI, which accounts for up to 10% of patients with HI, usually responds to diazoxide and, in KS, may resolve within a few years [47]. A series from a North American HI referral centre found HI more frequently than published cohorts and more neonatal intensive care admissions and feeding tubes in patients with HI, although the authors acknowledged selection bias [59].

3.12 Immunity and autoimmunity

Immune dysfunction combines humoral deficiency and autoimmune disease and may resemble common variable immunodeficiency [2]. Sinopulmonary infections and recurrent otitis occur in most patients; hypogammaglobulinaemia and IgA deficiency are common and B-lymphocyte populations may be reduced [2,39]. Vitiligo, immune thrombocytopenia, autoimmune haemolytic anaemia and diabetes mellitus have been described, particularly in childhood and adolescence [2]. In the UK paediatric immune thrombocytopenia registry (2006–2020), 5 of 2013 patients had KS, which, compared with a prevalence of 1:32,000, corresponds to an estimated relative prevalence of 79 (95% confidence interval 10–149); presentation and course resembled sporadic immune thrombocytopenia, and only one case became chronic, unlike the 12 of 13 chronic cases in an earlier French series [60].

3.13 Tumours

Malignant tumours have been described in isolated cases, but there is no clear evidence of significant cancer predisposition, and no tumour screening protocol has been developed [2,20]. In a Turkish series of 13 patients with KS1, one had pilomatricoma and another ganglioneuroblastoma [61].

3.14 Occasional manifestations

Smith records premature greying, vitiligo, cleft lip, lower lip pits, middle and inner ear malformations, microtia, cutaneous syndactyly, nail hypoplasia, cryptorchidism, micropenis, delayed puberty, imperforate anus, umbilical and inguinal hernias, precocious puberty, obesity, pectus excavatum, diaphragmatic hernia or eventration, biliary atresia, bronchial tree stenosis and growth hormone deficiency as occasional findings [20].

4. Clinical Recognition and Warning Signs

Recognition of KS depends on age. Prenatally, the 2010 European guideline states that suspicion may arise from nuchal oedema or associated malformations, although facies is inconspicuous in the newborn and, if uncertain, later reassessment is preferable [12]; in the KDM6A series, 37% of pregnancies with available information had abnormal findings, chiefly intrauterine growth restriction [3]. In newborns and infants, the combination of hypotonia, severe feeding difficulty, persistent hypoglycaemia, left-sided obstructive heart disease (aortic coarctation, left heart hypoplasia), cleft palate, diaphragmatic hernia or eventration, anorectal malformation or renal anomaly should prompt consideration of KS, particularly because HI precedes KS diagnosis in around 60% of HI cases with KS [2,25,52]. The Oxford Desk Reference includes KS among diagnoses to consider with long palpebral fissures, a heart murmur in a patient with ptosis, congenital heart disease (aortic coarctation), cleft palate and neonatal hypoglycaemia [27,48,58,62]. From the first year onwards, facies becomes characteristic, alongside fetal fingertip pads, recurrent otitis, premature thelarche in girls, downward deviation of the growth curve and developmental and language delay [2,20,27]. In adolescence and adulthood, truncal obesity, patellar dislocation and autoimmune cytopenias may dominate the clinical picture [2,12].

Some signs require immediate action. Hypoglycaemia, even in a child already diagnosed with KS, should be confirmed and investigated with a critical sample, because HI may present even after the first year [2,25]. A seizure requires exclusion of hypoglycaemia as a precipitating factor [12]. Recurrent lower respiratory tract infections should prompt consideration, besides immunodeficiency, of aspiration, bronchial tree anomalies and diaphragmatic eventration, and prolonged jaundice should prompt exclusion of biliary atresia [12]. Purpura, pallor or signs of haemolysis suggest autoimmune cytopenia [2,12]. Nocturnal lagophthalmos requires corneal protection [2,12]. Sensorineural hearing loss should prompt petrous bone imaging, because Mondini dysplasia with a perilymphatic fistula requires surgery to prevent meningitis [12]. Computer-assisted facial analysis may support recognition: in a Chinese cohort, Face2Gene correctly suggested KS in 85 of 99 children, and the authors attributed the remaining cases to poor photograph quality; patients with atypical facies were diagnosed later and had a more diverse and severe phenotype [49].

5. Diagnosis: Criteria, Tests and Algorithms

5.1 2019 international consensus criteria

The consensus criteria were drawn up by an international panel of clinical and molecular experts from Canada, the European Union, Japan, the United Kingdom and the USA, meeting for one day in Boston, with teleconference links to Japan and England, within the Kabuki Syndrome Medical Advisory Board organised and sponsored by Takeda in January 2018, which, according to the authors, did not influence the content; the article was published online in December 2018 and in the 2019 volume of the Journal of Medical Genetics [11]. The systematic PubMed search (terms “Kabuki”, “Kabuki make-up” and “Niikawa-Kuroki”) excluded articles predating the identification of KMT2D, those not specific to KS, those limited to the cellular function of the genes and those not published in English; of the 123 articles read in full text, 53 were excluded (aggregate data only, pooled KMT2D and KDM6A data, or previously published patients), leaving 70 publications with 449 molecularly confirmed patients (Table 3.1b) [11]. The panel discussed whether to adopt a phenotypic score or a system of major and minor criteria, like that of tuberous sclerosis complex, sought to balance the expected sensitivity and specificity, and decided not to use a score [11]. The need for clinical criteria was justified by four limitations of molecular diagnosis: sequencing has an analytical sensitivity of about 99% but does not detect all causal variants; the genes are large (42,921 base pairs in KMT2D and 239,605 in KDM6A); deep intronic variants or variants in regulatory elements escape clinical methods; and testing is not universally available or may be equivocal, as with a variant of uncertain significance [11].

A definitive diagnosis of KS can be made at any age, in either sex, in an individual with a history of infantile hypotonia, developmental delay and/or intellectual disability and one or both major criteria: a pathogenic or likely pathogenic variant in KMT2D (heterozygous) or in KDM6A (heterozygous or hemizygous); or typical dysmorphic features at some point in life [2,11]. The variant must be pathogenic or likely pathogenic according to the ACMG/AMP (American College of Medical Genetics and Genomics/Association for Molecular Pathology) criteria of Richards and colleagues; a variant of uncertain significance cannot be used as a major criterion [11]. Hypotonia and developmental delay replace intellectual disability in the mandatory criterion because the latter cannot be assessed in the newborn and infant and not all patients have it [11]. Table 5.1 details the definition of the dysmorphic features.

Those who do not meet the major criteria may have a probable or possible diagnosis, defined using a table of supportive clinical findings (Table 5.1b) [11]. A probable diagnosis requires a history of infantile hypotonia, developmental delay and/or intellectual disability, long palpebral fissures (measurement 2 SD or more above the mean for age) with eversion of the lateral third of the lower eyelid, and at least three supportive findings, in any combination of systems (one from each of three systems, three from the same system or another combination); the presence of other typical dysmorphic features increases the likelihood of the diagnosis, as does a Makrythanasis score of 6.0 or more [11]. A possible diagnosis requires a history of developmental delay and/or intellectual disability, at least two supportive findings and at least one of the four accessory features of the clinical major criterion (eyebrows, columella, ears or fingertip pads) at some point in life [11]. For probable and possible diagnoses, the authors recommend seriously considering clinical genetic testing, because these features are not specific to KS, and possibly other investigations, such as chromosomal microarray, to look for conditions with overlapping features [11]. The transcription published by Cerqueira and colleagues in 2025 matches the original text; the 2021 summary by Boniel and colleagues describes the probable diagnosis as infantile hypotonia and developmental delay with at least three supportive findings and the possible diagnosis as two findings, omitting the palpebral fissure requirement in the former and the dysmorphic feature requirement in the latter, and therefore diverges from the consensus [11,35,63]. In practice, the probable diagnosis corresponds to the patient with the typical periocular facies without enough accessory features for the major criterion, and the possible diagnosis to the patient with accessory features without the periocular facies; in both cases the classification guides the request for molecular testing and the methylation signature, and does not replace them [2,11].

The choice of supportive findings favoured specificity. Blue sclerae, although more frequent in KMT2D variant carriers, occur in several connective tissue disorders and are a normal finding before 1 year of age; premature thelarche (isolated breast development in a girl younger than 8 years), rare in other genetic syndromes, occurs in up to 90% of normal newborns; neither was included [11]. Long palpebral fissures with eversion of the lateral third of the lower eyelid were considered relatively specific, although not pathognomonic, and are therefore required for a definitive or probable clinical diagnosis; the remaining facial features are less specific [11]. The panel incorporated, in different combinations, the ethnicity-independent minimal clinical criteria proposed by Cheon and colleagues (long palpebral fissures with lateral eversion of the lower eyelid, short columella with depressed nasal tip, broad arched eyebrows with lateral sparseness, cupped or prominent ears and developmental delay or cognitive impairment) [11]. Applied to three published patients with a mosaic KMT2D variant and a mild phenotype, the clinical criteria, ignoring the molecular result, would have given a definitive diagnosis in two; the third, with fewer published phenotypic data, met only the molecular criterion [11,32]. The authors acknowledge that the sensitivity and specificity of the criteria have not been evaluated and that they may need revision, particularly as the KS2 phenotype becomes better known, since the frequency of its features in large cohorts was unknown at the time [11].

The phenotypic score of Makrythanasis and colleagues (Table 5.1c) was designed to predict the presence of a KMT2D variant, not as a diagnostic criterion; applied prospectively to 86 clinically diagnosed patients, it gave a mean of 6.1 in patients with a KMT2D variant and 4.5 in the others, a statistically significant difference [11]. In two small Czech validation studies with overlapping patients, the mean was 8.0 (7.0–9.0) in the 6 of 14 patients with a KMT2D variant and 4.8 (4.0–7.0) in patients without a variant in either gene, with a score of 7.0 assigned to both one positive and one negative patient; in the second study (24 patients, 7 of whom had been included in the first), 10 of the 15 patients (66.7%) with a score above 6.0 had a KMT2D variant and no patient with a score below 5.0 had one [11,64]. The consensus accepts the score as support for the clinical diagnosis in patients with a probable or possible diagnosis [11].

Table 5.1c — Phenotypic score of Makrythanasis and colleagues (as adapted in the 2019 consensus, Table 1)

Source: [11]. Note: the table reproduces the adaptation published in the consensus; the definition of each feature follows the original article by Makrythanasis and colleagues.

Table 5.1b — Probable and possible diagnosis criteria and supportive clinical findings (2019 consensus, Table 2)

Table 5.1 — Criteria for definitive KS diagnosis (2019 consensus)

These criteria replaced previous clinical criteria. The 2010 European guideline, predating gene identification, applied criteria derived from Adam and Hudgins (2005): characteristic facies, postnatal growth retardation, developmental delay or learning difficulty and, as accessory criteria, hypodontia and fingertip pads; it also required exclusion of chromosomal abnormalities by microarray [12]. A practical limitation of current criteria is that facies is atypical in more than one third of patients with KDM6A, so the clinical criterion underestimates KS2 [3,4]; another is that patients with the allelic syndrome involving KMT2D exons 38–39 may formally meet the molecular criterion without having KS [2].

5.2 Molecular testing strategy

GeneReviews proposes two approaches. When the phenotype suggests KS, single-gene testing or a multigene panel may be used: sequence KMT2D first, followed by KMT2D deletion/duplication analysis if negative, and then KDM6A sequencing and deletion/duplication analysis; a panel including KMT2D, KDM6A and differential diagnosis genes, with deletion/duplication analysis, is the recommended option to limit variants of uncertain significance in genes that do not explain the phenotype [2]. Since patients with a classical phenotype and a mosaic KMT2D variant have been described, targeted next-generation sequencing is preferable to Sanger sequencing [2,32]. When the phenotype is atypical and KS has not been considered, exome (or genome) sequencing is the best option [2]. The international HI guideline recommends genetic testing in all children with HI, except when an acquired form is likely (strong recommendation, moderate evidence), and assessment for multisystem syndromes in all infants with HI (strong recommendation, low evidence); it lists KMT2D and KDM6A among syndromic HI genes with a diffuse histological pattern [13]. Rosenfeld and colleagues recommend including both genes in the genetic assessment of all HI cases, because facies may not be recognisable in infancy [25].

Table 5.2 — Performance of molecular methods (GeneReviews, Table 1)

Source: [2]. Note: the "proportion" column concerns the fraction of KS cases explained by the gene; subsequent columns indicate the fraction of pathogenic variants in that gene detected by each method, rather than clinical test sensitivity in a patient with suspected KS.

5.3 DNA methylation signature (episignature)

KS has a genomic DNA methylation signature in peripheral blood leucocytes. In 24 patients with pathogenic KMT2D variants and 216 controls, analysis of more than 450,000 CpG sites allowed construction of a methylation-based pathogenicity index with 100% sensitivity and specificity, confirmed in multiple databases, and reclassification of variants of uncertain significance in patients with a suggestive phenotype [24]. The KS1 signature differs from the CHARGE syndrome signature and controls [41]. In an assessment of 42 Mendelian neurodevelopmental disorders, 34 had robust episignatures, and the authors showed that classification using a single signature at a time may misclassify closely related signatures, requiring multiclass modelling [65]. GeneReviews recommends considering episignature analysis in blood or stored blood-derived DNA in two situations: clinical suspicion of KS without a pathogenic KMT2D or KDM6A variant after sequencing or genomic testing; and clinical suspicion with a variant of uncertain significance in either gene [2].

The signature also validates mosaicism. In an Italian study, the signature correctly classified 13 patients with KMT2D variants of uncertain significance: it confirmed diagnosis in 3 and rejected the pathogenic role of 10 variants, in 3 of whom exome sequencing found another cause; in 5 patients with postzygotic variants, the authors identified 15% as the minimum mosaicism threshold above which the test is informative [66]. A North American case of severe hypoplastic left heart syndrome with a de novo KMT2D nonsense variant present in 11.2% of reads was confirmed by the EpiSign platform, suggesting a lower blood mosaicism limit compatible with diagnosis [34]. In 2025, a positive signature in a patient without a molecular diagnosis led to data reanalysis and identification of an AluY insertion in KMT2D [33]. In May 2026, a next-generation KS1 episignature was published, constructed using 110 patients with KMT2D variants and 854 controls, through microarrays and long-read sequencing: the classifier achieved 97% sensitivity and 100% specificity in validation cohorts, showed better concordance with clinical findings than in silico tools, maintained performance across platforms, reflected mosaicism levels and classified pathogenic KDM6A variants as positive [43]. An observational study of episignatures in rare prenatal-onset diseases is recruiting in France (NCT06475651) [67].

5.4 Test limitations and report content

Sequencing detects single-nucleotide variants and small insertions/deletions, but usually does not detect exon or whole-gene deletions or duplications, which require targeted analysis (quantitative PCR, long-range PCR, multiplex ligation-dependent probe amplification, targeted microarray) [2]. A negative result does not exclude KS with a convincing phenotype: around 30% of clinical diagnoses remain unexplained, low-level mosaicism may evade Sanger sequencing, and rare mechanisms such as mobile element insertions may not be recognised by standard pipelines [2,33,34]. GeneReviews recommends considering DNA storage from probands without a molecular diagnosis [2]. The report should specify the gene, versioned transcript (NM_003482.4 for KMT2D; for KDM6A, MANE Select NM_001291415.2 and, ideally, NM_021140, used in the literature), HGVS nomenclature at coding DNA and protein levels, genomic coordinate with genome version, zygosity (heterozygous, hemizygous or mosaic, with allele fraction), the five-class ACMG/AMP classification and method limitations [2,5]. A variant of uncertain significance does not become pathogenic through clinical response or partial segregation; parental investigation and the episignature are reclassification tools, and a missense variant in KMT2D exons 38–39 should be interpreted in light of the distinct allelic syndrome [2,26,66].

5.5 Biochemical diagnosis of hyperinsulinism

GeneReviews recommends considering a baseline blood glucose measurement at all ages and, in newborns and infants with persistent hypoglycaemia, investigating HI with a "critical sample" (insulin, free fatty acids, beta-hydroxybutyrate and glycaemic response to glucagon during hypoglycaemia) [2]. The 2024 international guideline recommends basing diagnosis on measurements obtained during hypoglycaemia (strong recommendation, moderate evidence) and defines the criteria presented in Table 5.3 [13].

Table 5.3 — Biochemical HI criteria during hypoglycaemia (plasma glucose <50 mg/dL [2.8 mmol/L])

Source: [13]. The diagnostic fasting test should be performed in an experienced unit, with venous access and dextrose available, measuring glucose and beta-hydroxybutyrate at defined intervals and collecting the critical sample when plasma glucose is ≤50 mg/dL; the glucagon test (30 µg/kg or 0.5–1 mg intramuscularly or intravenously) follows the critical sample [13].

5.6 Prenatal diagnosis

Once the pathogenic variant has been identified in an affected relative, prenatal diagnosis and preimplantation genetic testing are possible [2]. Without a known variant, prenatal diagnosis depends on ultrasound detection of non-specific malformations (heart disease, renal anomalies, diaphragmatic hernia) and fetal exome or genome sequencing; prenatal findings described in KS2 include intrauterine growth restriction [3,12].

5.7 Assessments after diagnosis

Table 5.4 — Recommended assessments after diagnosis

5.8 Diagnostic algorithm

In practice, the pathway can be described in five steps, based on the sources cited. First, in an infant with hypotonia, feeding difficulty and one or more suggestive malformations, or with HI, include KS in the differential even without the typical facies [2,25]. Second, request a panel with KMT2D and KDM6A that includes deletion/duplication analysis, or an exome/genome if the phenotype is atypical; if mosaicism is suspected, prefer next-generation sequencing with adequate depth [2]. Third, if the result is positive, check whether the KMT2D variant lies in exons 38–39 and whether the phenotype is compatible with KS or with the allelic syndrome [2,26]. Fourth, if the result is negative or shows a variant of uncertain significance with a suggestive phenotype, request the episignature and, if positive, reanalyse the data for structural, intronic or mobile-element variants [2,33]. Fifth, if the episignature is negative, review the differential diagnosis, including Kabuki phenocopies and chromosomal microarray [2,20], following the sequence in Table 6.3.

6. Differential Diagnosis

6.1 Entities with clinical overlap

The differential diagnosis of KS is organised into four groups: syndromes with overlapping malformations (CHARGE, 22q11.2 deletion, branchio-oto-renal syndrome, IRF6-related disorders), connective tissue disorders with hypermobility, chromosomal anomalies and Kabuki phenocopies caused by other genes. Table 6.1 brings together the entities with the features that distinguish them; phenocopies, or “Kabuki-like” phenotypes, are developed in sections 6.2 to 6.4.

Table 6.1 — Differential diagnosis

In practice, CHARGE is the most frequently discussed differential. Patients with KS may have coloboma, choanal atresia or ear anomalies, and a 2025 Brazilian report describes an adult with a previous phenotypic CHARGE syndrome diagnosis in whom a renal anomaly panel (including CHD7) was negative and a second panel identified a de novo KMT2D exon 38 variant (see section 10) [35]. The Williams syndrome differential table in the Brazilian genetic syndromes textbook includes KS because of shared congenital heart disease and developmental delay [74].

6.2 “Kabuki-like” phenotypes: definition and proportion

A “Kabuki-like” phenotype, or Kabuki phenocopy, designates the picture of a patient with a clinical diagnosis of KS, or with features suggesting it, in whom testing of KMT2D and KDM6A does not identify a pathogenic variant [14]. The term is descriptive and does not designate a nosological entity: it groups patients with true KS not detected by the methods applied (low-level mosaicism, deletions and duplications, deep intronic variants or mobile-element insertions) and patients with other syndromes whose facies overlaps with that of KS [2,14,33,75]. GeneReviews does not use the designation; it reports that about 30% of clinical diagnoses of KS remain without an identified cause and cites RAP1A, RAP1B and KDM6C as candidate or overlapping genes [2,44].

The proportion of patients without a variant in KMT2D or KDM6A depends on the rigour of the clinical diagnosis and the methods used, and none of the series measures a population frequency. The 2018 review by Lintas and Persico estimates that patients without a variant in either gene account for about 30% of KS cases, with 56–75% attributed to KMT2D and 3–8% to KDM6A [14]. In a Turkish cohort of 37 children aged 1 to 16 years, followed for a median of 6.8 years, sequencing and multiplex ligation-dependent probe amplification (MLPA) of both genes in 32 patients identified the cause in 22 (68.7%; 21 in KMT2D and 1 in KDM6A); exome sequencing of the remainder revealed no pathogenic variant, and the frequency of facial features, heart defects, renal anomalies, short stature, microcephaly and epilepsy did not differ between patients with truncating variants, with non-truncating variants and without a variant [76]. In a North American series of 27 probands with a clinical diagnosis of KS studied by targeted sequencing, 12 had a causal variant in the two known genes and 2 had presumed loss-of-function de novo variants in KMT2A, the Wiedemann-Steiner syndrome gene [77]. Conversely, the typical facies predicts the presence of a KMT2D variant: in the cohort of 116 patients of Banka and colleagues, almost all patients with the typical facies had a pathogenic variant [28]. In a Czech study, lower values of the “MLL2-Kabuki” phenotypic score were associated with “Kabuki-like” phenotypes, a group in which chromosomal microarray and sequencing of a Mendelian disease gene panel had a high diagnostic yield [64].

6.3 Overlapping genes and syndromes

Table 6.2 brings together the entities described in patients initially classified as KS or “Kabuki-like”, with the inheritance, the features that distinguish them from KS and the discriminating test.

Table 6.2 — “Kabuki-like” phenotypes: genes, syndromes and discriminating tests

Au-Kline syndrome is the best-characterised phenocopy. Okamoto syndrome, described in 1997, is now considered the same condition; the 2024 GeneReviews reports more than 75 patients known to the authors, more than 40 published, all with a de novo variant [15,78]. The clinical criteria of Choufani and colleagues (2022) require at least four of six facial features (long palpebral fissures, ptosis, shallow orbits, deeply grooved tongue, broad nose with broad nasal root and thick alae, down-turned mouth with M-shaped upper lip) and global developmental delay or intellectual disability with a quotient below 70 [15,79]. The authors applied these criteria to 22 patients with KS confirmed by a variant in KMT2D or KDM6A, assessed by three specialists, to test their specificity [79]. According to GeneReviews, arched and broad eyebrows, a short columella with depressed nasal tip, cleft lip or lip pits and immunodeficiency point to KS, and craniosynostosis is rare in KS [15]. HNRNPK loss-of-function variants are associated with moderate to severe intellectual disability, and missense variants with a sometimes more favourable development [15]. Sequencing detects 70 of 75 variants and deletion/duplication analysis 5 of 75; five deep intronic variants were detected only by genome sequencing, and contiguous gene deletions at 9q21.3 add manifestations of neighbouring genes [15]. The HNRNPK methylation signature, built with nine patients with loss-of-function variants, classified ten pathogenic KMT2D variants as “non-Au-Kline” and made it possible to classify missense variants in patients with a mild phenotype [79]. In a British patient with a “Kabuki-like” phenotype and nodular heterotopia, the HNRNPK variant was identified by exome sequencing in March 2012, but its significance was recognised only after the publication, in August 2015, of two similar patients found through the GeneMatcher platform [69]. Dentici and colleagues propose considering this diagnosis particularly in the presence of nodular heterotopia, craniosynostosis or polydactyly [70].

The overlap with Wiedemann-Steiner syndrome has a functional basis: KMT2A, like KMT2D, encodes a histone H3 lysine 4 (H3K4) methyltransferase [80]. The KMT2A episignature was built in a cohort of 60 patients with a clinical diagnosis of Wiedemann-Steiner syndrome or KS; Kabuki1 samples separated completely from Wiedemann-Steiner samples, and the four samples without a known KMT2A variant, with a Wiedemann-Steiner or KS phenotype, clustered with the Wiedemann-Steiner samples [80]. CHARGE syndrome is covered in section 6.1.

The link between RAP1A and RAP1B and KS rests on two Turkish patients described in 2015 by Bögershausen and colleagues. The first, with a KS picture, was homozygous for the RAP1A variant NM_002884.4:c.488G>C p.(Arg163Thr) (described in the original article as NM_001010935.2:c.488G>C), through paternal uniparental isodisomy of chromosome 1; the second, with a “Kabuki-like” phenotype and an initial clinical diagnosis of Hadziselimovic syndrome, had the de novo RAP1B variant NM_001010942.3:c.451A>G p.(Lys151Glu) (NM_015646.6:c.451A>G in the article) [5,81–83]. The authors showed in zebrafish and in fibroblasts that dysfunction of KMT2D, KDM6A, RAP1A and RAP1B disrupts MEK/ERK signalling and actin polymerisation, that MEK inhibitors corrected these defects in zebrafish and that KS shares mechanisms with the RASopathies; these data are preclinical [82]. Sequencing of RAP1A in 78 KS patients without a KMT2D variant identified no other pathogenic variant [82]. OMIM classifies the RAP1B variant as of uncertain significance and associates gain-of-function RAP1B variants (Gly12, Ala59, Gly60) with a syndromic thrombocytopenia with multiple congenital anomalies and facial dysmorphism (THC11), considered a RASopathy [83].

Terminal 2q37 deletions enter the differential because of facial dysmorphism and developmental delay. In a Spanish study of 16 patients with KS or a “Kabuki-like” phenotype studied by microarray, two had complex 2q rearrangements with a terminal 2q37 deletion (an 11 Mb inverted duplication with a 4.5 Mb terminal deletion and a 7.2 Mb terminal deletion), both de novo; the study did not confirm the 8p23.1-p22 duplication previously proposed as a cause of KS nor find any other recurrent rearrangement [84]. In 2q37 deletion syndrome, type E brachydactyly is present in about half of patients, the critical gene is HDAC4 and there are families with an inherited deletion and normal cognition [16]. HDAC4 missense variants cause a distinct entity (NEDCHF), with long palpebral fissures in five of six patients assessed [85].

On the X chromosome, a 13-month-old child with mosaic Turner syndrome (60% 45,X cells and the remainder 46,XX with a ring marker chromosome), a mild “Kabuki-like” phenotype and mild hyperinsulinism responded to diazoxide [86]; the hypothesis that KDM6A haploinsufficiency contributes to these pictures is mentioned in section 2.7 [44,47]. An Xp11.3 microdeletion including KDM6A and NDP caused Norrie disease and KS2 in a male newborn; the mother had a clinical diagnosis of KS and asymptomatic peripheral retinal vascular changes [87]. Three partial KDM6A duplications detected by targeted sequencing in newborns with hyperinsulinism without the typical facies were initially classified as of uncertain significance; genome sequencing showed that one disrupted the reading frame (pathogenic), and the KS episignature was positive in one of the two remaining patients and negative in the other [75].

The remaining phenocopies come from isolated cases or small series: biallelic OTUD6B variants in three siblings, with an initial suspicion of KS also in previously published patients [88]; a de novo MED13 variant in a 13-year-old girl with a prior clinical diagnosis of KS [90]; Temtamy syndrome (C12orf57) in an Indian child with eversion of the lateral third of the lower eyelid and fetal fingertip pads [92]; a late-onset Zellweger spectrum disorder [94]; and, in series of “Kabuki-like” patients without a variant in either gene, a pathogenic CNV at 14q11.2, variants in HUWE1, GRIN1 and EFTUD2 and a 5 Mb de novo deletion at 10p11.22-p11.21 [14,64]. Barry and colleagues also include ZMIZ1 among the genes implicated in KS-like phenotypes [71]. In the functional enrichment analysis of Lintas and Persico, genes associated with “Kabuki-like” phenotypes converged on chromatin remodelling and transcriptional regulation [14].

6.4 Genetic approach to the patient with a “Kabuki-like” phenotype

The approach starts from two checks: whether testing of KMT2D and KDM6A was complete (sequencing with sufficient depth for mosaicism and deletion/duplication analysis of both genes, bearing in mind that about 20% of KDM6A variants are deletions or duplications) and whether the phenotype meets the KS criteria or those of another entity, such as Au-Kline [2,15,79]. Table 6.3 summarises the testing sequence described in the sources; the order may change with the phenotype and local access to tests.

Table 6.3 — Testing sequence in the “Kabuki-like” phenotype

The methylation signature is the tool that best separates KS from its phenocopies. In the set of 57 episignatures described by Levy and colleagues in 2022, KS1 and KS2 form a common class (KMT2D and KDM6A), built with 149 samples [96]; the 2026 next-generation KS1 episignature also classified pathogenic KDM6A variants as positive [43]. The Au-Kline, CHARGE and Wiedemann-Steiner signatures are distinct from the KS signature [41,79,80]. In a German cohort of 400 patients with neurodevelopmental disorders studied by exome and episignature, 67 (17%) had variants in chromatinopathy genes; of the 60 with a methylation profile, 26 (43%) had a specific episignature concordant with the clinical diagnosis, and 21 of the variants were in genes without a signature in the panel used [97]. A negative episignature therefore does not exclude a chromatinopathy in a gene without a validated signature. Conversely, a KS signature in a variant-negative patient should prompt a search for rare mechanisms in KMT2D and KDM6A, as shown by the retrospective identification of an AluY insertion in KMT2D [33].

7. Treatments and Risk Management Strategies

7.1 General principles and recommendation sources

There is no treatment correcting the epigenetic defect; management is symptomatic, multidisciplinary and directed at affected organs [2,4]. Available recommendations have three origins: the GeneReviews treatment and surveillance tables, subject to focused revisions through April 2026 [2]; the 2010 European guideline, developed by the DYSCERNE network with methodology adapted from the Scottish Intercollegiate Guidelines Network for rare diseases, predominantly based on expert consensus and described by GeneReviews as not updated [2,12]; and, for HI, the 2024 international guideline, grading recommendation strength and evidence quality [13]. No randomised trial of any KS intervention was identified [4]. The European guideline recommends referring all families to a genetics service and identifying a person responsible for care coordination [12].

7.2 Hyperinsulinism

Acute treatment of neonatal hypoglycaemia due to HI is with intravenous glucose: a bolus of 200 mg/kg (2 mL/kg of 10% dextrose) followed by a continuous infusion sufficient to keep plasma glucose above 70 mg/dL, typically 8 mg/kg/min or more, increasing in increments of 4 mg/kg/min or more if necessary, with no maximum dose defined in the source and avoiding fluid overload (strong recommendation, moderate evidence); if intravenous access is unavailable, glucagon may be used at a pharmacological dose (0.5–1 mg or 20–30 µg/kg, intramuscularly or subcutaneously, with an effect lasting 40–60 minutes), and a continuous glucagon infusion (2.5–20 µg/kg/h) may reduce fluid intake when glucose requirements are high [13]. The therapeutic target is to keep plasma glucose between 70 and 100 mg/dL (3.9–5.6 mmol/L) [13].

Diazoxide, a benzothiadiazine derivative that opens ATP-sensitive potassium channels of the beta cell and suppresses insulin secretion, is the first-line treatment (strong recommendation, moderate evidence) [13]. In the United States, Proglycem® (oral suspension 50 mg/mL) has a recommended starting dose in newborns and infants of 8 mg/kg/day, divided into three doses every 8 hours or two doses every 12 hours, titratable to a maximum of 15 mg/kg/day; in children and adults, the starting dose is 3 mg/kg/day, titratable to 8 mg/kg/day, and refractory hypoglycaemia may require higher doses; the label recommends discontinuing the drug if it is not effective after 2 to 3 weeks [98]. The international guideline indicates a therapeutic range of 5–15 mg/kg/day in infants and children, with no additional benefit and more adverse effects above this range, and recommends starting a diuretic at the same time, particularly in the newborn and with doses above 10 mg/kg/day: chlorothiazide 10 mg/kg/day or hydrochlorothiazide 1–2 mg/kg/day, with no maximum dose defined in the source [13]. Response is demonstrated when the child can fast and generate hyperketonaemia (beta-hydroxybutyrate >1.8 mmol/L) before glucose falls below 50–60 mg/dL; the drug should be discontinued if the dextrose infusion cannot be reduced after 5 days at 15 mg/kg/day or if lack of response is confirmed on a fasting test [13].

Acute adverse effects include sodium and water retention, with fluid overload, oedema, hyponatraemia, tachypnoea and respiratory failure, and pulmonary hypertension (2.4% in a large infant cohort); in the long term, hypertrichosis is most common (84.1%), followed by coarsening of facial features (24%), neutropenia (15.6%), thrombocytopenia (4.7%) and hyperuricaemia (5.0%); serious events requiring drug discontinuation occur in around 9.7% [13]. The guideline recommends echocardiography one week after initiation, and a full blood count with differential and uric acid every 6 months [13]. The US label adds that fluid retention may precipitate heart failure in patients with compromised cardiac reserve, that ketoacidosis and non-ketotic hyperosmolar coma may occur particularly during intercurrent illness, and describes postmarketing reversible pulmonary hypertension in newborns and infants, pericardial effusion and necrotising enterocolitis, the latter chiefly in infants with comorbidities [98]. These warnings are particularly relevant in KS because around 70% of patients have heart disease, frequently left-sided obstructive disease [10]. In the Philadelphia cohort, the centre always combined a diuretic (usually chlorothiazide) with diazoxide and requested cardiological assessment in children with heart disease; diazoxide was not tried in 4 children with significant left ventricular outflow tract malformations; the median initial dose was 8.4 mg/kg/day (interquartile range 7.3–13.3); among responders, one patient had fluid overload without pulmonary hypertension and another persistent neutropenia and anaemia, the latter switched to lanreotide 60 mg monthly at 16 months; non-responders or children without a diazoxide trial received continuous intragastric dextrose (median rate 4.0 mg/kg/min) or frequent or continuous enteral feeding [25]. In the French series, the three children with persistent HI responded to diazoxide monotherapy, and one subsequently required additional octreotide [42].

Somatostatin analogues (octreotide, lanreotide) are suggested as second line in diazoxide non-responders, in those with unacceptable adverse effects or when diazoxide is unavailable (conditional recommendation, low evidence); the guideline does not establish a dose for long-acting formulations, owing to insufficient data [13]. The guideline describes diazoxide as the only drug currently approved to treat HI, which makes the use of somatostatin analogues in this indication off-label; the monthly lanreotide 60 mg mentioned above is a cohort observation and not a dosing recommendation [13,25]. The guideline advises against drugs without proof of efficacy, such as nifedipine and sirolimus, outside research protocols [13]. HI in KS tends to resolve: in the Philadelphia cohort, treatment was stopped in 46% at a median age of 2.8 years and, among carriers of KDM6A variants, in all five [25]. Fasting tolerance should be assessed before discharge, and all children with HI should be referred for developmental surveillance, feeding assessment and early intervention (strong recommendation) [13].

7.3 Growth and endocrine system

GeneReviews states that short stature, even without growth hormone deficiency, responded to growth hormone without worsening disproportion: after one year of treatment, mean height changed from −2.40 to −1.69 standard deviations, and benefit was greater with earlier treatment initiation [2]. Conversely, Ruault and colleagues concluded that in their French cohort, growth hormone did not increase height beyond that predicted [54]. The European guideline recommends endocrinology referral if height or weight is below the 2nd percentile or growth velocity decreases, measuring growth hormone if height is below the 3rd percentile and monitoring scoliosis during treatment [12]. In the US somatropin labels consulted on openFDA (Genotropin®, Norditropin®, Omnitrope®, Zomacton®, Humatrope®), paediatric indications include growth hormone deficiency and several syndromes, but not KS; treatment of children with KS without documented deficiency is therefore off-label under these labels, and no KS-specific dose is available in the sources consulted [99]. Hypothyroidism and HI are treated by the endocrinologist; isolated premature thelarche requires no treatment without other signs of precocious puberty [2]. The European guideline recommends monitoring puberty (possibly later in boys), advising diet and exercise from puberty because of the tendency towards obesity and, for menstrual management, allowing girls to progress normally through puberty before considering combined oral contraceptives or injectable medroxyprogesterone acetate (Depo-Provera®), without routine endometrial ablation or hysterectomy [12].

7.4 Heart

Treatment of heart defects and arrhythmias is standard; whether aortic aneurysm risk is increased is unclear, but this potential risk should be communicated to the team before catheterisation or angioplasty [2,12]. The European guideline recommends annual blood pressure measurement in all patients and annual surveillance of aortic dilatation in those with aortic anomalies, because of dissection risk, and Digilio and colleagues recommend echocardiography at diagnosis and annual surveillance of arch dilatation [10,12]. Antibiotic prophylaxis before and during procedures (for example, dental procedures) may be indicated in specific heart conditions [2]. Surgical outcomes depend on anatomical complexity: at a North American centre, seven newborns with KS and left-sided lesions operated on between 2018 and 2023 had hospital survival of 71% (versus 88% in newborns with other genetic abnormalities), more unplanned interventions (43% versus 15%), more brain imaging abnormalities (29% versus 5%), more bacteraemia (29% versus 9%), more ventilation days (median 16 versus 6) and longer admissions (66 versus 41 days) [100]; at another centre, between 2000 and 2025, 11 of 18 patients with heart disease underwent surgery and 4 (36%) died during the index admission, predominantly with single-ventricle physiology, without late mortality in survivors or non-operated patients [50]. These data come from small single-centre series and do not allow estimation of individual risks, but justify multidisciplinary perioperative planning.

7.5 Immunity and autoimmunity

Intravenous immunoglobulin may be considered in patients with documented immunoglobulin deficiency, with referral to immunology [2]. The European guideline permits the normal vaccination schedule when immune values are normal and recommends immunology advice about vaccination if abnormal [12]. In the British registry, KS-associated immune thrombocytopenia had a presentation and course resembling the sporadic form and is treated in the same way [60]. Patients with autoimmune disease should be referred to immunology [12].

7.6 Sensory organs, face and teeth

Strabismus, refractive errors, ptosis and lagophthalmos receive standard ophthalmological treatment; nocturnal lagophthalmos may require lubrication to protect the cornea, and recurrent eye infections from lacrimal pathway stenosis or atresia justify ophthalmology or oculoplastic referral [2,12]. Conductive hearing loss due to chronic otitis media is treated with tympanostomy tubes and sensorineural loss with hearing aids; cochlear implantation may be considered, and at least one patient with profound progressive sensorineural loss has been described with improved quality of life after implantation [2]. The European guideline adds speech therapy, surgery for ossicular malformations and, in Mondini dysplasia with a perilymphatic fistula, surgery to prevent meningitis [12]. Cleft lip and palate should be treated by a specialist team, recognising that the palate may be shorter and velopharyngeal insufficiency after repair more frequent; lip pits are excised only if problematic [2,12]. Hypodontia and malocclusion justify orthodontic or paediatric dental care; the European guideline recommends an early preventive oral health programme and states that dental caries are not more frequent [2,12].

7.7 Feeding, gastrointestinal tract and kidney

Gastro-oesophageal reflux is treated by thickening milk, positioning after meals and, if necessary, medicines; severe difficulties or poor suck–swallow coordination may require gastrostomy [2]. Chronic diarrhoea warrants gastroenterological assessment for malabsorption and coeliac disease; constipation, attributed to hypotonia, is treated with standard laxatives [2,12]. Urological anomalies (hypospadias, cryptorchidism) receive standard urological treatment [2]. Given the frequency of chronic kidney disease in the Korean cohort, the authors recommend screening for renal and urinary anomalies and regular renal function monitoring [51].

7.8 Neurology, development and behaviour

Seizures are treated with standard antiseizure medicines [2]. The European guideline recommends physiotherapy from diagnosis for hypotonia and discourages invasive examinations, such as muscle biopsy, to investigate other causes of hypotonia [12]. GeneReviews recommends early intervention from 0 to 3 years (occupational therapy, physiotherapy, speech and feeding therapy), adapted education with an individual plan, neurodevelopmental therapies focused on language and motor skills, and initiation of transition planning for adulthood from 12 years [2]. Children may benefit from interventions used in autism spectrum disorder, such as applied behaviour analysis, and medication for attention deficit hyperactivity disorder when indicated [2]. In adolescence and adulthood, the European guideline notes that many patients acquire basic literacy and recommends continuing speech therapy and training in social, independence and employment skills [12].

7.9 Musculoskeletal system

The European guideline recommends physiotherapy and occupational therapy for hyperlaxity, hip surveillance until walking (dislocation may arise even with normal neonatal screening), annual clinical spinal examination with radiography if scoliosis is suspected, and annual assessment of symptoms and patellar position [12]. The Portuguese case of severe thoracic kyphosis described in section 10 illustrates proximal junctional kyphosis risk after arthrodesis and the interaction between ligamentous hyperlaxity, extent of correction and postoperative infection [101].

7.10 Anaesthesia

The anaesthetist should know the diagnosis and associated malformations; heart disease should be screened for before surgery, difficult intubation anticipated due to structural airway anomalies, and the cervical spine positioned carefully because of hyperlaxity; cleft palate repair should be performed by experienced anaesthetists at a cleft centre [2,12].

7.11 Agents and circumstances to avoid

In patients with hyperlaxity, activities increasing joint injury risk, such as trampolining, should be avoided; the European guideline adds avoidance of bouncy seats in infants [2,12].

7.12 Medicines used and regulatory status

Table 7.1 — Regulatory status of relevant medicines (accessed 6 October 2026)

No information on reimbursement or specific access in Portugal (INFARMED) or Brazil (ANVISA/CONITEC) for these medicines in KS was found in the sources consulted.

7.13 Investigational therapies and clinical trials

The rationale for investigational therapies is the model that KS results from a reversible imbalance between open and closed chromatin. In Kmt2d+/βGeo mice, the histone deacetylase inhibitor AR-42, the ketogenic diet and the LSD1/KDM1A inhibitor TAK-418 corrected hippocampal neurogenesis and memory [36–38]. AR-42 also partially normalised DNA methylation at sites of the murine blood signature, suggesting utility as a response biomarker in future trials [107]. GeneReviews states that neither deacetylase inhibitors nor the ketogenic diet are recommended as treatment outside a clinical trial, and that some families and physicians have independently tried the diet without results published in peer-reviewed journals [2].

Table 7.2 — Registered or published human studies

These studies do not allow conclusions about efficacy. The modified Atkins diet study has ten participants, no control group (a limitation stated in the registry), multiple primary outcomes without correction or published statistical analysis, unmasked assessment and possible learning effects between test administrations; means for one measure (brief visuospatial memory) appear in the registry with group labels inconsistent with the scale ranges and were not interpreted in this review [108]. Human TAK-418 studies were phase 1 in healthy volunteers and included no patients with KS; the European orphan designation was withdrawn and no trial registration in patients was found [105,109,110]. Searching the European clinical trials registry (CTIS) for "Kabuki" yielded no results on the access date. Positive results in mouse models do not equate to clinical benefit.

7.14 Developments over the last 12 months

Between October 2025 and October 2026, the following developments occurred, ordered by evidence level. Peer-reviewed publications: the next-generation KS1 episignature (May 2026; sensitivity 97%, specificity 100%, long-read sequencing assessment and fine mapping of KMT2D variant effects) [43]; the French series of 18 patients with neonatal hypoglycaemia in 33% and HI in 28% (January 2026) [42]; 25-year cardiological and surgical follow-up at a North American centre (September 2026) [50]; characterisation of the social phenotype, with more frequent autistic traits in KS2 (2026) [55]; and a KS1 missense variant mouse model with perinatal lethality and renal agenesis (September 2026) [31]. Reference sources: focused GeneReviews revisions concerning HI (March 2025) and episignature use in reclassifying variants of uncertain significance (April 2026) [2]; ClinGen reassessment of KMT2D clinical validity (August 2025) [6]. Registries: publication of modified Atkins diet study results (May 2025, slightly before the 12-month window) and initiation of the French prenatal episignature study (February 2026) [67,108]. No regulatory opinions, authorisation applications or authorisations for any KS-specific therapy at the EMA or FDA were identified on the access date.

8. Reproductive Options and Family Counselling

8.1 KS1 (KMT2D, autosomal dominant)

KS1 is inherited in an autosomal dominant manner. A minority of probands have an affected parent; the proportion of de novo cases has not been rigorously determined because simplex cases have not been systematically studied, but clinical experience indicates it is high [2]. In the founding molecular-era study, the variant was de novo in 12 families and transmitted in two, always consistent with the phenotype [21], and the 1988 Japanese series contained only sporadic cases [8]. Parental testing for the variant identified in the proband is recommended; if absent from both parents' leucocyte DNA, the variant is probably de novo, but parental germline mosaicism, although not described, is possible [2]. If the proband's variant is unknown, parents should be examined by a clinical geneticist, because a mildly affected parent may have escaped diagnosis, and a parent in whom the variant arose may have somatic mosaicism with a minimal phenotype [2]. Sibling risk is 50% if a parent is affected; if the variant is not detected in the parents, risk is presumed slightly above that of the general population (still below 1%) because of the theoretical possibility of germline mosaicism [2]. Each child of a person with KS1 has a 50% chance of inheriting the variant [2]. When neither parent has the variant, non-medical explanations should also be considered, such as alternative paternity or maternity (including assisted reproduction) or undisclosed adoption [2].

8.2 KS2 (KDM6A, X-linked)

The father of an affected boy is not a carrier and does not require testing [2]. In a family with more than one affected individual, the mother of an affected boy is an obligate heterozygote; if a woman has more than one affected child without other affected relatives and the variant is not detected in her leucocyte DNA, germline mosaicism is the most likely explanation [2]. Around 80% of affected boys are simplex cases [2]. In the Faundes series, de novo variants were more frequent in girls (92.1%) than boys (62.5%), and when inheritance information was complete, inherited variants accounted for 59% of boys and 42.9% of protein-altering non-truncating variants; one boy inherited a mosaic loss-of-function variant from an unaffected mother [3]. If the mother has the variant, transmission risk is 50% in each pregnancy: boys inheriting it will be affected, and girls will be heterozygous and may have KS traits, usually milder [2]. Maternal aunts and their descendants may be at risk [2].

Sources diverge concerning affected men: GeneReviews states that no affected man has been described as having children [2], whereas Faundes and colleagues describe a girl inheriting the p.Ala149Thr variant (NM_021140.3 numbering) from an equally affected father, the first documented paternal transmission, and show that some women with KS2 have children [3]. Published experience with offspring of people with KS2 is, in any case, very limited [2,3].

8.3 Reproductive planning

The ideal time to determine genetic risk and discuss prenatal diagnosis and preimplantation genetic testing is before pregnancy, and counselling should be offered to young adults who are affected or carriers [2]. Once the pathogenic variant has been identified in an affected relative, prenatal diagnosis and preimplantation genetic testing are possible [2]. In cases without a molecular diagnosis, DNA banking of the proband should be considered [2]. No data on pregnancy in women with KS, on fertility or on obstetric outcome were found in the sources consulted; the European guideline addresses only contraception and menstrual management [12]. In patients with a “Kabuki-like” phenotype without a molecular diagnosis, the sources consulted do not allow a specific recurrence risk, which depends on the cause: in Au-Kline syndrome all described probands had a de novo variant and each child of an affected person has a 50% probability of inheriting the variant; 2q37 deletions may be inherited from a parent with a mild phenotype or normal cognition, which justifies testing the parents; and the RAP1A variant was inherited from an unaffected heterozygous father and made homozygous by uniparental isodisomy [15,16,81]. DNA banking and periodic reanalysis of genomic data keep open the possibility of a later diagnosis [2,69].

9. Special Considerations: Population Diversity and Epidemiology

9.1 Frequency

The most cited estimate comes from the 1988 Japanese study: 1:32,000 newborns, calculated in a series in which 58 of 62 patients were Japanese, leading the authors to consider the syndrome relatively frequent in Japan [8]. GeneReviews accepts that birth prevalence outside Japan approaches the Japanese figure [2]. In Australia and New Zealand, White and colleagues calculated a minimum birth prevalence of 1:86,000, with mean age at diagnosis of 3 years and 10 months, although facies was evident from childhood in most patients [9]; the minimum nature of this estimate and diagnostic delay explain part of the difference. Orphanet classifies birth prevalence in the 1–9 per 100,000 category, with the Japanese value of 3.1 per 100,000 [4]. For the 2018 European orphan designation, the EMA recorded, based on sponsor information, a prevalence of around 1 in 10,000 in the European Union; this is an estimate for regulatory purposes, not measured in a population study [105]. These frequencies apply to KS of all causes; there is no measured KS2 frequency, and the GeneGuidance figure of around 0.13 per 100,000 is derived rather than observed [4]. No founder effect was described in the sources consulted, consistent with the high proportion of de novo variants.

9.2 American, European and other population data

In the United States, recent cohorts come predominantly from referral centres: the Philadelphia HI cohort (33 children; 55% described as White, 9% Asian and 6% Hispanic) [25], the general cohort at the same hospital, with acknowledged HI selection bias [59], cardiac surgical series [50,100] and the modified Atkins diet study (nine White participants and one Asian participant) [108]. In Europe, prominent data include the British series of 116 patients [28], the multicentre KDM6A series centred on the United Kingdom, Belgium and United States [3], French growth charts [54], the French KS network (110 patients recruited between 2008 and 2015) [112], Italian heart disease and episignature series [10,66], the Dutch growth series [53] and the British immune thrombocytopenia registry [60]. In Asia, founding Japanese data, the Korean kidney disease cohort and the Chinese cohort of 104 children are available, the latter with lower heart disease frequency (27%) than European series and anomalous pulmonary venous connection as a new finding [8,49,51]. These cohorts measure frequencies in patients followed at specialist centres and should not be converted into population prevalences.

9.3 Portugal and Brazil

No Portuguese or Brazilian epidemiological data were found in the sources consulted. Case reports exist: an adolescent followed in Lisbon with severe thoracic kyphosis and proximal junctional kyphosis after arthrodesis [101], and a São Paulo adult with a KMT2D exon 38 variant and a previous CHARGE syndrome diagnosis [35], both described in section 10. The Brazilian genetic syndromes textbook includes KS in the Williams syndrome differential diagnosis [74]. Absence of data in the sources consulted does not demonstrate that no studies exist in these countries.

9.4 Special populations

In KS2, sex influences the phenotype, with greater severity in boys [3]. For adults, the 2010 European guideline is the only consulted source with specific recommendations (obesity, menstrual management, contraception in vulnerable women, independent living skills) [12]. In patients with somatic mosaicism, the phenotype may be milder, with fewer heart defects, and diagnosis may require deep sequencing and an episignature [32,34]. Patients with complex heart disease form the group with the greatest perioperative mortality [50,100].

9.5 European reference networks and patient associations

KS falls within ERN ITHACA (European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders), which made an educational video about KS available in May 2026 and issued a collaboration call for the allelic syndrome involving KMT2D exons 38–39 [17,45]; searching the network website did not identify a dedicated KS guideline. Patient associations listed by GeneReviews are All Things Kabuki, the Kabuki Syndrome Foundation (United States) and SAKKS (Supporting Aussie Kids with Kabuki Syndrome, Australia) [2]; the 2010 European guideline also listed the Kabuki Syndrome Network (Canada), Netwerk Kabuki Syndroom (the Netherlands) and Association Syndrome de Kabuki (France), whose contact details may have changed since then [12].

10. Illustrative Clinical Case

10.1 Severe thoracic kyphosis and proximal junctional kyphosis in a Portuguese adolescent with KS1

Nóbrega and colleagues, from the Orthopaedics Department of Unidade Local de Saúde de São José in Lisbon, described in 2026 a 15-year-old boy with genetically confirmed KS through a KMT2D variant (the article does not specify the variant), with a history of congenital heart disease, dysplastic aorta, attention deficit hyperactivity disorder and multiple joint dislocations, referred for progressive thoracic deformity and low back pain [101]. He had rigid thoracic kyphosis without scoliosis or neurological deficit, with Risser 4 skeletal maturity; the lateral radiograph showed thoracic kyphosis with a Cobb angle of 95° (Figure 1 in the article: https://pmc.ncbi.nlm.nih.gov/articles/PMC12801274/figure/F1/) and magnetic resonance imaging was normal [101]. He underwent posterior instrumentation and arthrodesis (pedicle screws at T5, T7, T11, T12 and L1, transverse process hooks at T2–T3, Smith-Petersen osteotomies from T6 to T10, cantilever technique), reducing Cobb angle to 60° [101]. At 8 months, proximal junctional kyphosis developed above the uppermost instrumented vertebra (Figure 3: https://pmc.ncbi.nlm.nih.gov/articles/PMC12801274/figure/F3/); at 11 months, fixation was extended to the cervicothoracic junction (C5–T5) and distal instrumentation removed [101]. Three weeks later, he developed a postoperative methicillin-sensitive Staphylococcus aureus infection, treated with intravenous flucloxacillin for 3 weeks, debridement and wound revision, followed by 10 weeks of oral co-trimoxazole and rifampicin [101]. Despite resolution of the infection, kyphosis progressed to 134°, a Milwaukee brace was applied and, six months after revision, halo-gravity traction was used for 3 weeks (5 to 10 kg); definitive surgery consisted of instrumentation revision from C5 to L3, Ponte osteotomies from T5 to T11 and repeat posterolateral arthrodesis [101]. At 12 months after revision, arthrodesis was consolidated with a Cobb angle of 100° and no further junctional deterioration; at 2 years, the patient had better posture, less pain and had resumed swimming; Figure 6 in the article shows widening of the scar, which the authors attribute to connective tissue laxity [101].

The authors interpret junctional kyphosis as multifactorial (severe preoperative kyphosis, extensive correction, a rigid proximal transition, infection and, possibly, KS ligamentous hyperlaxity and immune dysfunction), but acknowledge that systematic reviews do not identify syndromic hyperlaxity as an independent predictor and that data on perioperative infection risk in KS are lacking [101]. Proposed preventive measures (preservation of the soft tissue envelope at the upper vertebra, careful upper-level selection, rod contouring, preoperative immune assessment) are the authors' opinions based on one case rather than guideline recommendations. The case illustrates three management points: annual clinical spinal surveillance recommended by the European guideline, the need for perioperative planning integrating heart disease, laxity and immunity, and the value of prior immune assessment [2,12,101].

10.2 Brazilian adult with a previous CHARGE syndrome diagnosis and a KMT2D exon 38 variant

Cerqueira and colleagues, from São Paulo, described in 2025 a 24-year-old man referred to nephrology for declining renal function [35]. He was born prematurely with congenital hypothyroidism, atrophic nipples, low-set ears, hypospadias, choanal atresia and a surgically treated branchial fistula, and had received a phenotypic CHARGE syndrome diagnosis; bilateral vesicoureteral reflux and hypospadias were corrected at 2 years, and he had dental abnormalities (small anterior teeth, triangular canines, implants) and severe sensorineural hearing loss requiring hearing aids from 3 years [35]. At 15 years, he had a hypocalcaemic crisis (ionised calcium 0.69 mmol/L, parathyroid hormone 20.3 pg/mL with reference range 20–65, vitamin D 6.1 ng/mL, phosphate 7.7 mg/dL), interpreted as primary hypoparathyroidism, and neuropsychological assessment revealed mild intellectual disability [35]. Estimated glomerular filtration rate was 76.2 mL/min/1.73 m² at 18 years and creatinine clearance 59.2 mL/min/1.73 m² at 24 years, with right kidney function of 25% on DMSA scintigraphy [35]. A congenital kidney and urinary tract anomaly panel including CHD7 was negative; a second panel identified the heterozygous NM_003482.4:c.10595T>C p.(Ile3532Thr) variant in KMT2D exon 38, absent from the parents, which the laboratory reclassified from uncertain significance to likely pathogenic, and the authors concluded that this was a form of KS distinct from typical KS [35].

This patient's phenotype (athelia, choanal atresia, branchial anomaly, hearing loss, hypothyroidism, hypoparathyroidism, mild intellectual disability, without description of typical facies) and variant location (residue 3532, within the region bounded by Leu3525 and Lys3583) correspond to the allelic syndrome involving exons 38–39 described by Baldridge and Cuvertino, which GeneReviews considers distinct from KS despite potentially formally meeting consensus criteria [2,5,26,30]. The "Kabuki" designation used by the authors therefore differs from current classification. The case also shows that variant reclassification relied on phenotypic correlation and de novo status, and that the authors themselves acknowledge the absence of functional studies; methylation analysis, which distinguishes this entity from KS1, is not mentioned in the report [26,35]. The article contains no clinical figures.

11. Prognosis and Long-Term Follow-up

11.1 Prognosis

The sources consulted provide no life expectancy estimates for KS. Mortality is concentrated in patients with complex heart disease, particularly single-ventricle disease, in the perioperative period [50,100]; in the Australian and New Zealand series, 3 of 27 patients had died [9]. Most patients speak and walk, and intellectual disability is usually mild to moderate in KS1 and more severe in boys with KS2 [2,3]. HI tends to resolve in the early years [25], but unrecognised hypoglycaemia may cause irreversible neurological injury [2]; in congenital HI generally, Saudubray describes bilateral cavitary occipital lesions on magnetic resonance imaging following profound prolonged hypoglycaemia with coma or status epilepticus in the newborn [47]. Adolescent and adult obesity, chronic kidney disease (one third of patients in a Korean cohort) and autoimmune cytopenias are long-term complications [2,51,60]. There is no clear evidence of significant cancer predisposition [2]. Epigenetic therapeutic hypotheses, demonstrated in mouse models, have not yet changed human prognosis [2,38].

11.2 Surveillance

Table 11.1 — Recommended surveillance

Tumour surveillance is not indicated because no screening protocol has been developed and there is no evidence of significant predisposition [2]. At-risk relatives should be assessed as outlined in section 8 [2].

12. Summary and Final Considerations

KS is a chromatinopathy with multiple congenital anomalies and a neurodevelopmental disorder, described in 1981 by Niikawa and Kuroki, caused by loss of function of KMT2D (KS1, autosomal dominant, about 75% of cases) or KDM6A (KS2, X-linked, 3–5%), with about 30% of clinical diagnoses remaining without an identified cause [2,18,19]. A definitive diagnosis, according to the 2019 consensus, rests on infantile hypotonia or developmental delay associated with a pathogenic variant and/or the typical facies, but the facies appears late in the newborn and is atypical in more than one third of patients with KS2 [3,11]. Testing should include sequencing and deletion/duplication analysis of both genes; the DNA methylation signature resolves variants of uncertain significance, confirms mosaicism above about 15% and, since 2026, has a next-generation version with 97% sensitivity and 100% specificity [2,43,66]. KDM6A numbering in the literature (NM_021140) does not match the MANE Select transcript (NM_001291415.2), and missense variants in exons 38–39 of KMT2D define a distinct allelic syndrome [5,26]. In patients with a “Kabuki-like” phenotype, the diagnosis depends on review of testing of both genes, microarray, trio exome or genome and multiclass episignature; the most frequently described causes are Au-Kline syndrome (HNRNPK), CHARGE syndrome, Wiedemann-Steiner syndrome, RAP1A/RAP1B variants, 2q37 deletions and X-chromosome anomalies [15,79,80,82,84,97].

Management is multidisciplinary and based on GeneReviews, the 2010 European guideline and, for HI, the 2024 international guideline [2,12,13]. HI occurs in both genotypes, manifests on the first day of life in most cases but is diagnosed late, responds to diazoxide in more than 90%, and frequently resolves in the early years; the high frequency of heart disease requires adding a diuretic and assessing the heart before and during treatment [13,25,98]. Left-sided obstructive heart defects, humoral immunodeficiency and autoimmunity, hearing loss, chronic kidney disease and joint hyperlaxity require structured surveillance [2,10,51]. No targeted therapy is approved; deacetylase inhibitors, the ketogenic diet and LSD1 inhibitors have shown effects in mouse models, the only patient study (modified Atkins diet, ten adults, uncontrolled) does not allow efficacy conclusions, and the European TAK-418 orphan designation was withdrawn [36–38,105,108]. Principal gaps are the absence of population prevalence studies outside Japan and Oceania, lack of data on adult life and reproduction, and absence of an updated management guideline.

13. References

1. Orphanet. Kabuki syndrome (ORPHA:2322): definition, synonyms and ICD-10/ICD-11 coding [database; Orphacodes API, version of 2 July 2026]. Paris: INSERM; accessed on 2026 Oct 6. Available at: https://www.orpha.net/en/disease/detail/2322
2. Adam MP, Hannibal M. Kabuki syndrome. 2011 Sep 1 [updated 2019 Feb 28; revision 2026 Apr 23]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. PMID: 21882399. Available at: https://www.ncbi.nlm.nih.gov/books/NBK62111/
3. Faundes V, Goh S, Akilapa R, Bezuidenhout H, Bjornsson HT, Bradley L, et al. Clinical delineation, sex differences, and genotype-phenotype correlation in pathogenic KDM6A variants causing X-linked Kabuki syndrome type 2. Genet Med. 2021;23(7):1202-10. doi:10.1038/s41436-021-01119-8. PMID: 33674768.
4. GeneGuidance. Gene and condition curation: KMT2D, KDM6A; Kabuki syndrome 1 and 2 [database]. Version of 26 September 2026; accessed on 2026 Oct 6. Additional query on 2026 Oct 29: CHD7, KMT2A, HNRNPK, HDAC4, OTUD6B, PEX1, HUWE1, GRIN1, EFTUD2.
5. 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):61-8. doi:10.1002/humu.23348. PMID: 28967166. Accessed on 2026 Oct 6 and 2026 Nov 1 (GRCh38). Available at: https://variantvalidator.org
6. Clinical Genome Resource (ClinGen). Gene-disease validity: KMT2D–Kabuki syndrome 1 (MONDO:0007843; Definitive, 2025 Aug 14) and KDM6A–Kabuki syndrome 2 (MONDO:0010465; Definitive, 2021 Jul 26); Dosage sensitivity: KMT2D and KDM6A [database]. Accessed on 2026 Oct 6. Available at: https://search.clinicalgenome.org
7. Banka S, Lederer D, Benoit V, Jenkins E, Howard E, Bunstone S, et al. Novel KDM6A (UTX) mutations and a clinical and molecular review of the X-linked Kabuki syndrome (KS2). Clin Genet. 2015;87(3):252-8. doi:10.1111/cge.12363. PMID: 24527667.
8. Niikawa N, Kuroki Y, Kajii T, Matsuura N, Ishikiriyama S, Tonoki H, et al. Kabuki make-up (Niikawa-Kuroki) syndrome: a study of 62 patients. Am J Med Genet. 1988;31(3):565-89. doi:10.1002/ajmg.1320310312. PMID: 3067577.
9. White SM, Thompson EM, Kidd A, Savarirayan R, Turner A, Amor D, et al. Growth, behavior, and clinical findings in 27 patients with Kabuki (Niikawa-Kuroki) syndrome. Am J Med Genet A. 2004;127A(2):118-27. doi:10.1002/ajmg.a.20674. PMID: 15108197.
10. Digilio MC, Gnazzo M, Lepri F, Dentici ML, Pisaneschi E, Baban A, et al. Congenital heart defects in molecularly proven Kabuki syndrome patients. Am J Med Genet A. 2017;173(11):2912-22. doi:10.1002/ajmg.a.38417. PMID: 28884922.
11. Adam MP, Banka S, Bjornsson HT, Bodamer O, Chudley AE, Harris J, et al. Kabuki syndrome: international consensus diagnostic criteria. J Med Genet. 2019;56(2):89-95. doi:10.1136/jmedgenet-2018-105625. PMID: 30514738.
12. Kabuki Syndrome Guideline Development Group (Clayton-Smith J, Philip N, Devriendt K, Digilio MC, Genevieve D, MacKinnon R, et al.). Management of Kabuki syndrome: a clinical guideline. Version 1. Manchester: DYSCERNE – A Network of Centres of Expertise in Dysmorphology, University of Manchester; 2010 Mar 8 [accessed on 2026 Oct 6]. Available at: https://kr.ihc.com/ext/Dcmnt?ncid=521096055&tfrm=default
13. De Leon DD, Arnoux JB, Banerjee I, Bergada I, Bhatti T, Conwell LS, et al. International guidelines for the diagnosis and management of hyperinsulinism. Horm Res Paediatr. 2024;97(3):279-98. doi:10.1159/000531766. PMID: 37454648.
14. Lintas C, Persico AM. Unraveling molecular pathways shared by Kabuki and Kabuki-like syndromes. Clin Genet. 2018;94(3-4):283-295. doi:10.1111/cge.12983. PMID: 28139835.
15. Au PYB, McNiven V, Phillips L, Innes AM, Kline AD. Au-Kline syndrome. 2019 Apr 18 [updated 2024 Feb 1]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available at: https://www.ncbi.nlm.nih.gov/books/NBK540283/
16. Online Mendelian Inheritance in Man (OMIM). #600430 Chromosome 2q37 deletion syndrome [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/600430
17. European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders (ERN ITHACA). Institutional website and educational documentation (Kabuki syndrome, video published on 2026 May 19) [Internet]. Accessed on 2026 Oct 6. Available at: https://ern-ithaca.eu/education/documentation-for-patients/kabuki-syndrome/
18. Niikawa N, Matsuura N, Fukushima Y, Ohsawa T, Kajii T. Kabuki make-up syndrome: a syndrome of mental retardation, unusual facies, large and protruding ears, and postnatal growth deficiency. J Pediatr. 1981;99(4):565-9. doi:10.1016/s0022-3476(81)80255-7. PMID: 7277096.
19. Kuroki Y, Suzuki Y, Chyo H, Hata A, Matsui I. A new malformation syndrome of long palpebral fissures, large ears, depressed nasal tip, and skeletal anomalies associated with postnatal dwarfism and mental retardation. J Pediatr. 1981;99(4):570-3. doi:10.1016/s0022-3476(81)80256-9. PMID: 7277097.
20. Jones KL, Jones MC, del Campo M. Kabuki syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 138-41.
21. Ng SB, Bigham AW, Buckingham KJ, Hannibal MC, McMillin MJ, Gildersleeve HI, et al. Exome sequencing identifies MLL2 mutations as a cause of Kabuki syndrome. Nat Genet. 2010;42(9):790-3. doi:10.1038/ng.646. PMID: 20711175.
22. Lederer D, Grisart B, Digilio MC, Benoit V, Crespin M, Ghariani SC, et al. Deletion of KDM6A, a histone demethylase interacting with MLL2, in three patients with Kabuki syndrome. Am J Hum Genet. 2012;90(1):119-24. doi:10.1016/j.ajhg.2011.11.021. PMID: 22197486.
23. Miyake N, Mizuno S, Okamoto N, Ohashi H, Shiina M, Ogata K, et al. KDM6A point mutations cause Kabuki syndrome. Hum Mutat. 2013;34(1):108-10. doi:10.1002/humu.22229. PMID: 23076834.
24. Aref-Eshghi E, Schenkel LC, Lin H, Skinner C, Ainsworth P, Paré G, et al. The defining DNA methylation signature of Kabuki syndrome enables functional assessment of genetic variants of unknown clinical significance. Epigenetics. 2017;12(11):923-33. doi:10.1080/15592294.2017.1381807. PMID: 28933623.
25. Rosenfeld E, Mitteer LM, Boodhansingh K, Sanders VR, McKnight H, De Leon DD. Clinical and molecular characterization of hyperinsulinism in Kabuki syndrome. J Endocr Soc. 2024;8(7):bvae101. doi:10.1210/jendso/bvae101. PMID: 38859884.
26. Cuvertino S, Hartill V, Colyer A, Garner T, Nair N, Al-Gazali L, et al. A restricted spectrum of missense KMT2D variants cause a multiple malformations disorder distinct from Kabuki syndrome. Genet Med. 2020;22(5):867-77. doi:10.1038/s41436-019-0743-3. PMID: 31949313.
27. Firth HV, Hurst JA. Ptosis, blepharophimosis, and other eyelid anomalies. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 292-4.
28. Banka S, Veeramachaneni R, Reardon W, Howard E, Bunstone S, Ragge N, et al. How genetically heterogeneous is Kabuki syndrome?: MLL2 testing in 116 patients, review and analyses of mutation and phenotypic spectrum. Eur J Hum Genet. 2012;20(4):381-8. doi:10.1038/ejhg.2011.220. PMID: 22126750.
29. Li Y, Bögershausen N, Alanay Y, Simsek Kiper PO, Plume N, Keupp K, et al. A mutation screen in patients with Kabuki syndrome. Hum Genet. 2011;130(6):715-24. doi:10.1007/s00439-011-1004-y. PMID: 21607748.
30. Baldridge D, Spillmann RC, Wegner DJ, Wambach JA, White FV, Sisco K, et al. Phenotypic expansion of KMT2D-related disorder: beyond Kabuki syndrome. Am J Med Genet A. 2020;182(5):1053-65. doi:10.1002/ajmg.a.61518. PMID: 32083401.
31. Halldorsdottir ST, Vinod M, Gunnlaugsson HO, Bjornsdottir ED, Jacques Lasseur RM, Luperchio T, et al. A patient-derived missense mouse model of Kabuki syndrome 1. Dis Model Mech. 2026; published online ahead of print. doi:10.1242/dmm.052775. PMID: 42713652.
32. Lepri FR, Cocciadiferro D, Augello B, Alfieri P, Pes V, Vancini A, et al. Clinical and neurobehavioral features of three novel Kabuki syndrome patients with mosaic KMT2D mutations and a review of literature. Int J Mol Sci. 2017;19(1):82. doi:10.3390/ijms19010082. PMID: 29283410.
33. Sabbagh Q, Ruiz-Pallares N, Rastin C, Puechberty J, Guignard T, Jeandel C, et al. Reverse genotyping: unveiling Alu element insertion as a new cause of Kabuki syndrome using DNA methylation signature. Clin Epigenetics. 2025;17(1):69. doi:10.1186/s13148-025-01879-z. PMID: 40301896.
34. Montano C, Britton JF, Harris JR, Kerkhof J, Barnes BT, Lee JA, et al. Genome-wide DNA methylation profiling confirms a case of low-level mosaic Kabuki syndrome 1. Am J Med Genet A. 2022;188(7):2217-25. doi:10.1002/ajmg.a.62754. PMID: 35384273.
35. Cerqueira BP, Milhomem EA, Matos ACC, Pietrobom IG, Leprevost CM, Heilberg IP. Kabuki and CHARGE syndromes: overlapping symptoms and diagnostic challenges. Einstein (Sao Paulo). 2025;23:eRC1142. doi:10.31744/einstein_journal/2025RC1142. PMID: 39969027.
36. Bjornsson HT, Benjamin JS, Zhang L, Weissman J, Gerber EE, Chen YC, et al. Histone deacetylase inhibition rescues structural and functional brain deficits in a mouse model of Kabuki syndrome. Sci Transl Med. 2014;6(256):256ra135. doi:10.1126/scitranslmed.3009278. PMID: 25273096.
37. Benjamin JS, Pilarowski GO, Carosso GA, Zhang L, Huso DL, Goff LA, et al. A ketogenic diet rescues hippocampal memory defects in a mouse model of Kabuki syndrome. Proc Natl Acad Sci U S A. 2017;114(1):125-30. doi:10.1073/pnas.1611431114. PMID: 27999180.
38. Zhang L, Pilarowski G, Pich EM, Nakatani A, Dunlop J, Baba R, et al. Inhibition of KDM1A activity restores adult neurogenesis and improves hippocampal memory in a mouse model of Kabuki syndrome. Mol Ther Methods Clin Dev. 2021;20:779-91. doi:10.1016/j.omtm.2021.02.011. PMID: 33738331.
39. Lindsley AW, Saal HM, Burrow TA, Hopkin RJ, Shchelochkov O, Khandelwal P, et al. Defects of B-cell terminal differentiation in patients with type-1 Kabuki syndrome. J Allergy Clin Immunol. 2016;137(1):179-87.e10. doi:10.1016/j.jaci.2015.06.002. PMID: 26194542.
40. Pilarowski GO, Cazares T, Zhang L, Benjamin JS, Liu K, Jagannathan S, et al. Abnormal Peyer patch development and B-cell gut homing drive IgA deficiency in Kabuki syndrome. J Allergy Clin Immunol. 2020;145(3):982-92. doi:10.1016/j.jaci.2019.11.034. PMID: 31816409.
41. Butcher DT, Cytrynbaum C, Turinsky AL, Siu MT, Inbar-Feigenberg M, Mendoza-Londono R, et al. CHARGE and Kabuki syndromes: gene-specific DNA methylation signatures identify epigenetic mechanisms linking these clinically overlapping conditions. Am J Hum Genet. 2017;100(5):773-88. doi:10.1016/j.ajhg.2017.04.004. PMID: 28475860.
42. Gaudillière M, Armand T, Senée V, Villanueva C, Nicolino M, Perge K. Clinical presentation and molecular characteristics of Kabuki syndrome with congenital hyperinsulinism: a retrospective study. Cureus. 2026;18(1):e101532. doi:10.7759/cureus.101532. PMID: 41695002.
43. Wang M, Helal S, Torabi-Marashi A, Goodman S, Kallurkar P, Truong TK, et al. A next-generation episignature for Kabuki syndrome enables fine mapping of the impact of KMT2D variants to inform precision medicine. Am J Hum Genet. 2026;113(6):1233-52. doi:10.1016/j.ajhg.2026.04.010. PMID: 42134323.
44. Bögershausen N, Gatinois V, Riehmer V, Kayserili H, Becker J, Thoenes M, et al. Mutation update for Kabuki syndrome genes KMT2D and KDM6A and further delineation of X-linked Kabuki syndrome subtype 2. Hum Mutat. 2016;37(9):847-64. doi:10.1002/humu.23026. PMID: 27302555.
45. ERN ITHACA. Call for collaboration: variants in KMT2D (exons 38-39) – BCAHH syndrome (OMIM #620186) [Internet]. Nantes: CHU de Nantes; accessed on 2026 Oct 6. Available at: https://ern-ithaca.eu/our-research-activities/calls-for-collaboration/variants-in-kmt2d-exons-38-39-bcahh-syndrome/
46. Tekendo-Ngongang C, Kruszka P, Martinez AF, Muenke M. Novel heterozygous variants in KMT2D associated with holoprosencephaly. Clin Genet. 2019;96(3):266-70. doi:10.1111/cge.13598. PMID: 31282990.
47. Arnoux JB, de Lonlay P. Congenital hyperinsulinism and genetic disorders of insulin resistance and signalling. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 6, sections 6.1 (Clinical presentation) and 6.6 (Prognosis). doi:10.1007/978-3-662-63123-2_6.
48. Firth HV, Hurst JA. Congenital heart disease. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 112-5.
49. Wang Y, Hu F, Xu X, Tan J, Yu T, Li N, et al. Clinical delineation and genotype-phenotype correlation in 104 children with Kabuki syndrome: a single-center, cross-sectional and follow-up study in China. Eur J Pediatr. 2025;184(4):271. doi:10.1007/s00431-025-06103-x. PMID: 40146326.
50. Camacho M, Yerebakan C, McConnell P, Galantowicz M, Davidson CL, Alvarado C, et al. Congenital heart defects in Kabuki syndrome: perioperative and long-term outcomes from a 25-year single-center experience. World J Pediatr Congenit Heart Surg. 2026; published online ahead of print. doi:10.1177/21501351261477060. PMID: 42742415.
51. Han S, Lee H, Park PG, Choi N, Ahn YH, Ko JM, et al. Long-term kidney outcomes in patients with Kabuki syndrome. Pediatr Nephrol. 2025;40(10):3101-9. doi:10.1007/s00467-025-06815-0. PMID: 40437108.
52. Firth HV, Hurst JA. Congenital diaphragmatic hernia. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 715.
53. Schott DA, Blok MJ, Gerver WJ, Devriendt K, Zimmermann LJ, Stumpel CT. Growth pattern in Kabuki syndrome with a KMT2D mutation. Am J Med Genet A. 2016;170(12):3172-9. doi:10.1002/ajmg.a.37930. PMID: 27530205.
54. Ruault V, Corsini C, Duflos C, Akouete S, Georgescu V, Abaji M, et al. Growth charts in Kabuki syndrome 1. Am J Med Genet A. 2020;182(3):446-53. doi:10.1002/ajmg.a.61462. PMID: 31876365.
55. Ng R, Fanara P, Meier L, Harris J. The social behavioral phenotype of Kabuki syndrome. Arch Clin Neuropsychol. 2026;41(7):acag069. doi:10.1093/arclin/acag069. PMID: 42715378.
56. Schott DA, Gerver WJ, Stumpel CT. Growth hormone stimulation tests in children with Kabuki syndrome. Horm Res Paediatr. 2016;86(5):319-24. doi:10.1159/000449221. PMID: 27649541.
57. Yap KL, Johnson AEK, Fischer D, Kandikatla P, Deml J, Nelakuditi V, et al. Congenital hyperinsulinism as the presenting feature of Kabuki syndrome: clinical and molecular characterization of 9 affected individuals. Genet Med. 2019;21(1):233-42. doi:10.1038/s41436-018-0013-9. PMID: 29907798.
58. Firth HV, Hurst JA. Hypoglycaemia in the neonate and infant. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 182-4.
59. Dortenzio V, Barbaric L, Rosenfeld E, Dechene ET, Gilbert MA, Dulik MC, et al. Clinical characteristics of patients with Kabuki syndrome at a single tertiary children's hospital. Am J Med Genet A. 2025;197(6):e64003. doi:10.1002/ajmg.a.64003. PMID: 39936499.
60. Reynolds L, Williams B, Gurumurthy G, Grainger J. Immune thrombocytopenia in Kabuki syndrome, a comparison with non-Kabuki cases in the UK paediatric ITP registry. Orphanet J Rare Dis. 2025;20(1):249. doi:10.1186/s13023-025-03743-y. PMID: 40420286.
61. Doğan Arı AB, Türkyılmaz A, Çavdarlı B, Kılıç E. Clinical and molecular findings of KMT2D-related Kabuki syndrome: a series of 13 patients with 3 novel variants. Eur J Med Genet. 2025;78:105052. doi:10.1016/j.ejmg.2025.105052. PMID: 41061959.
62. Firth HV, Hurst JA. Cleft lip and palate. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 104-5.
63. Boniel S, Szymańska K, Śmigiel R, Szczałuba K. Kabuki syndrome—clinical review with molecular aspects. Genes (Basel). 2021;12(4):468. doi:10.3390/genes12040468. PMID: 33805950.
64. Paderova J, Drabova J, Holubova A, Vlckova M, Havlovicova M, Gregorova A, et al. Under the mask of Kabuki syndrome: Elucidation of genetic-and phenotypic heterogeneity in patients with Kabuki-like phenotype. Eur J Med Genet. 2018;61(6):315-321. doi:10.1016/j.ejmg.2018.01.005. PMID: 29307790.
65. Aref-Eshghi E, Kerkhof J, Pedro VP, Groupe DI France, Barat-Houari M, Ruiz-Pallares N, et al. Evaluation of DNA methylation episignatures for diagnosis and phenotype correlations in 42 mendelian neurodevelopmental disorders. Am J Hum Genet. 2020;106(3):356-70. doi:10.1016/j.ajhg.2020.01.019. PMID: 32109418.
66. Niceta M, Ciolfi A, Ferilli M, Pedace L, Cappelletti C, Nardini C, et al. DNA methylation profiling in Kabuki syndrome: reclassification of germline KMT2D VUS and sensitivity in validating postzygotic mosaicism. Eur J Hum Genet. 2024;32(7):819-26. doi:10.1038/s41431-024-01597-9. PMID: 38528056.
67. Assistance Publique – Hôpitaux de Paris. Characterization and contribution of genome-wide DNA methylation (DNA methylation episignatures) in rare diseases with prenatal onset. ClinicalTrials.gov identifier: NCT06475651 [Internet]. Updated on 2026 Mar 27; accessed on 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT06475651
68. Jones KL, Jones MC, del Campo M. Au-Kline syndrome (HNRNPK-related Kabuki syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 142-3.
69. Lange L, Pagnamenta AT, Lise S, Clasper S, Stewart H, Akha ES, et al. A de novo frameshift in HNRNPK causing a Kabuki-like syndrome with nodular heterotopia. Clin Genet. 2016;90(3):258-262. doi:10.1111/cge.12773. PMID: 26954065. PMC5006848.
70. Dentici ML, Barresi S, Niceta M, Pantaleoni F, Pizzi S, Dallapiccola B, et al. Clinical spectrum of Kabuki-like syndrome caused by HNRNPK haploinsufficiency. Clin Genet. 2018;93(2):401-407. doi:10.1111/cge.13029. PMID: 28374925.
71. Barry KK, Tsaparlis M, Hoffman D, Hartman D, Adam MP, Hung C, et al. From genotype to phenotype—a review of Kabuki syndrome. Genes (Basel). 2022;13(10):1761. doi:10.3390/genes13101761. PMID: 36292647.
72. Jones KL, Jones MC, del Campo M. Baraitser-Winter syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 314-6.
73. Clinical approach to inborn errors of metabolism in paediatrics. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, section 1.5 (table of cardiac symptoms). doi:10.1007/978-3-662-63123-2_1.
74. Ribeiro EM, coord. Síndrome de Williams (Box 2.6.3: differential diagnosis). In: Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 56-7.
75. Männistö JME, Hopkins JJ, Hewat TI, Nasser F, Burrage J, Dastamani A, et al. Congenital Hyperinsulinism and Novel KDM6A Duplications -Resolving Pathogenicity With Genome and Epigenetic Analyses. J Clin Endocrinol Metab. 2025;110(5):e1524-e1530. doi:10.1210/clinem/dgae524. PMID: 39078990. PMC12012811.
76. Usluer E, Sayın GY, Güneş N, Kasap B, Tüysüz B. Investigation of genetic and phenotypic heterogeneity in 37 Turkish patients with Kabuki and Kabuki-like phenotype. Am J Med Genet A. 2022;188(10):2976-2987. doi:10.1002/ajmg.a.62944. PMID: 36097644.
77. Sobreira N, Brucato M, Zhang L, Ladd-Acosta C, Ongaco C, Romm J, et al. Patients with a Kabuki syndrome phenotype demonstrate DNA methylation abnormalities. Eur J Hum Genet. 2017;25(12):1335-1344. doi:10.1038/s41431-017-0023-0. PMID: 29255178. PMC5865196.
78. Online Mendelian Inheritance in Man (OMIM). #616580 Au-Kline syndrome; AUKS [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/616580
79. Choufani S, McNiven V, Cytrynbaum C, Jangjoo M, Adam MP, Bjornsson HT, et al. An HNRNPK-specific DNA methylation signature makes sense of missense variants and expands the phenotypic spectrum of Au-Kline syndrome. Am J Hum Genet. 2022;109(10):1867-1884. doi:10.1016/j.ajhg.2022.08.014. PMID: 36130591. PMC9606382.
80. Foroutan A, Haghshenas S, Bhai P, Levy MA, Kerkhof J, McConkey H, et al. Clinical Utility of a Unique Genome-Wide DNA Methylation Signature for KMT2A-Related Syndrome. Int J Mol Sci. 2022;23(3):1815. doi:10.3390/ijms23031815. PMID: 35163737. PMC8836705.
81. Online Mendelian Inheritance in Man (OMIM). *179520 RAS-related protein 1A; RAP1A [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/179520
82. Bögershausen N, Tsai IC, Pohl E, Kiper PÖ, Beleggia F, Percin EF, et al. RAP1-mediated MEK/ERK pathway defects in Kabuki syndrome. J Clin Invest. 2015;125(9):3585-3599. doi:10.1172/jci80102. PMID: 26280580. PMC4588287.
83. Online Mendelian Inheritance in Man (OMIM). *179530 RAS-related protein 1B; RAP1B [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/179530
84. Cuscó I, del Campo M, Vilardell M, González E, Gener B, Galán E, et al. Array-CGH in patients with Kabuki-like phenotype: identification of two patients with complex rearrangements including 2q37 deletions and no other recurrent aberration. BMC Med Genet. 2008;9:27. doi:10.1186/1471-2350-9-27. PMID: 18405349. PMC2358878.
85. Online Mendelian Inheritance in Man (OMIM). #619797 Neurodevelopmental disorder with central hypotonia and dysmorphic facies; NEDCHF [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/619797
86. Pietzner V, Weigel JF, Wand D, Merkenschlager A, Bernhard MK. Low-level hyperinsulinism with hypoglycemic spells in an infant with mosaic Turner syndrome and mild Kabuki-like phenotype: a case report and review of the literature. J Pediatr Endocrinol Metab. 2014;27(1-2):165-170. doi:10.1515/jpem-2013-0090. PMID: 23950569.
87. Mansoor M, Coussa RG, Strampe MR, Larson SA, Russell JF. Xp11.3 microdeletion causing Norrie disease and X-linked Kabuki syndrome. Am J Ophthalmol Case Rep. 2023;29:101798. doi:10.1016/j.ajoc.2023.101798. PMID: 36703904. PMC9871737.
88. Gangaram B, Lee V, Slavotinek A. Biallelic OTUD6B variants associated with a Kabuki syndrome-like disorder in three siblings: A clinical report and literature review. Am J Med Genet A. 2024;194(7):e63567. doi:10.1002/ajmg.a.63567. PMID: 38389298.
89. Online Mendelian Inheritance in Man (OMIM). *603808 Mediator complex subunit 13; MED13 [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/603808
90. De Nardi L, Faletra F, D'Adamo AP, Bianco AMR, Athanasakis E, Bruno I, et al. Could the MED13 mutations manifest as a Kabuki-like syndrome? Am J Med Genet A. 2021;185(2):584-590. doi:10.1002/ajmg.a.61994. PMID: 33258286.
91. Online Mendelian Inheritance in Man (OMIM). #218340 Temtamy syndrome; TEMTYS [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/218340
92. Nerakh G, Vasikarla M. A novel pathogenic compound heterozygous variant in C12orf57 gene in a child with Temtamy syndrome presenting with overlapping phenotypic features of Kabuki-like syndrome. Brain Dev. 2023;45(9):512-516. doi:10.1016/j.braindev.2023.06.008. PMID: 37451886.
93. Online Mendelian Inheritance in Man (OMIM). *138450 Serine hydroxymethyltransferase, mitochondrial; SHMT2 [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/138450
94. Ezgu F, Eminoglu T, Okur I, Gunduz M, Tumer L, Hasanoglu A, et al. An infantile case of Zellweger syndrome presented with Kabuki-like phenotype. Genet Couns. 2011;22(2):217-220. PMID: 21848015.
95. Online Mendelian Inheritance in Man (OMIM). #618659 Neurodevelopmental disorder with dysmorphic facies and distal skeletal anomalies; NEDDFSA [Internet]. Baltimore (MD): Johns Hopkins University; accessed on 2026 Oct 29. Available at: https://www.omim.org/entry/618659
96. Levy MA, McConkey H, Kerkhof J, Barat-Houari M, Bargiacchi S, Biamino E, et al. Novel diagnostic DNA methylation episignatures expand and refine the epigenetic landscapes of Mendelian disorders. HGG Adv. 2022;3(1):100075. doi:10.1016/j.xhgg.2021.100075. PMID: 35047860. PMC8756545.
97. Koparir A, Kerkhof J, Rzasa J, Metzger E, Bahena Carbajal P, Kolokotronis K, et al. Chromatinopathies: clinically overlapping disorders, revealing novel variants and their DNA methylation signatures. Clin Epigenetics. 2026;18(1):69. doi:10.1186/s13148-026-02120-1. PMID: 41957673. PMC13088768.
98. Teva Pharmaceuticals USA. PROGLYCEM (diazoxide) oral suspension: prescribing information. Revision of July 2024 [SPL label, set id b16c7832-2fd9-49af-b923-1dc0d91fd6e2; accessed via openFDA on 2026 Oct 6]. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b16c7832-2fd9-49af-b923-1dc0d91fd6e2
99. U.S. Food and Drug Administration. Somatropin labels (GENOTROPIN, set id ffebf88b-d257-4542-9808-74d9b7167765, version of 2026 Aug 4; NORDITROPIN; OMNITROPE; ZOMACTON; HUMATROPE): Indications and Usage section [openFDA drug label API; accessed on 2026 Oct 6]. Available at: https://api.fda.gov/drug/label.json?search=openfda.generic_name:%22somatropin%22
100. Lam RW, Haro V, Tabbutt S, Kinami H, Reddy VM, Kim M. Kabuki patients with CHDs: outcomes after cardiac surgery. Cardiol Young. 2025;35(4):722-5. doi:10.1017/S1047951125000435. PMID: 40007435.
101. Nóbrega J, Almeida R, Rasteiro P, Rosado J, Botelho T, Carvalho N. Proximal junctional kyphosis following spinal thoracic deformity correction in a patient with Kabuki syndrome: a case report. J Orthop Case Rep. 2026;16(1):236-42. doi:10.13107/jocr.2026.v16.i01.6666. PMID: 41541477.
102. U.S. Food and Drug Administration. Drugs@FDA: NDA 017453 (PROGLYCEM), original approval on 1976 May 28 [openFDA drugsfda API; accessed on 2026 Oct 6]. Available at: https://api.fda.gov/drug/drugsfda.json?search=application_number:%22NDA017453%22
103. European Medicines Agency. Medicines data (human medicines): search for diazoxide and diazoxide choline (Viokat, application withdrawn) [Internet]. Accessed on 2026 Oct 6. Available at: https://www.ema.europa.eu/en/medicines/download-medicine-data
104. Soleno Therapeutics. VYKAT XR (diazoxide choline) extended-release tablets: prescribing information [SPL label, set id 0e745e85-9512-e360-e063-6394a90aebf1; NDA 216665 approved on 2025 Mar 26; accessed via openFDA on 2026 Oct 6]. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0e745e85-9512-e360-e063-6394a90aebf1
105. European Medicines Agency. EU/3/18/2082: orphan designation for the treatment of Kabuki syndrome (5-{(1R,2R)-2-[(cyclopropylmethyl)amino]cyclopropyl}-N-(tetrahydro-2H-pyran-4-yl)thiophene-3-carboxamide monohydrochloride; Takeda Pharma A/S) [Internet]. Designated 2018 Nov 19; withdrawn 2022 Nov; accessed on 2026 Oct 6. Available at: https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu-3-18-2082
106. National Center for Biotechnology Information. PubChem Compound Summary for CID 118432651, TAK-418 [Internet]. Bethesda (MD): National Library of Medicine; accessed on 2026 Oct 6. Available at: https://pubchem.ncbi.nlm.nih.gov/compound/118432651
107. Goodman SJ, Luperchio TR, Ellegood J, Chater-Diehl E, Lerch JP, Bjornsson HT, et al. Peripheral blood DNA methylation and neuroanatomical responses to HDACi treatment that rescues neurological deficits in a Kabuki syndrome mouse model. Clin Epigenetics. 2023;15(1):172. doi:10.1186/s13148-023-01582-x. PMID: 37884963.
108. Hugo W. Moser Research Institute at Kennedy Krieger, Inc. Study of modified Atkins diet in Kabuki syndrome. ClinicalTrials.gov identifier: NCT04722315 [Internet]. Results posted on 2025 May 13; accessed on 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT04722315
109. Yin W, Arkilo D, Khudyakov P, Hazel J, Gupta S, Quinton MS, et al. Safety, pharmacokinetics and pharmacodynamics of TAK-418, a novel inhibitor of the epigenetic modulator lysine-specific demethylase 1A. Br J Clin Pharmacol. 2021;87(12):4756-68. doi:10.1111/bcp.14912. PMID: 33990969.
110. ClinicalTrials.gov. Studies of TAK-418: NCT03228433 (completed), NCT03501069 (terminated: business decision; no human safety concern; evaluation of a preclinical toxicology finding) and NCT04202497 (terminated for administrative reasons) [Internet]. Accessed on 2026 Oct 6. Available at: https://clinicaltrials.gov/search?term=TAK-418
111. University Hospital, Montpellier. Exploiting epigenome editing in Kabuki syndrome: a new route towards gene therapy for rare genetic disorders. ClinicalTrials.gov identifier: NCT03855631 [Internet]. Accessed on 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT03855631
112. Assistance Publique – Hôpitaux de Paris. French Kabuki syndrome network: epidemiology, management of patients and research by array-CGH. ClinicalTrials.gov identifier: NCT01314534 [Internet]. Accessed on 2026 Oct 6. Available at: https://clinicaltrials.gov/study/NCT01314534