TAB2-Related Syndrome

Syndrome of cardiac anomalies, short stature and joint hypermobility due to TAB2 haploinsufficiency, including 6q25.1 microdeletions

Literature and regulatory information updated: 6 November 2026.

0. Initial Reference Table

1. Introduction

1.1. Definition

TAB2-related syndrome is an autosomal dominant genetic disorder caused by loss of function of one of the two alleles of the TAB2 gene, which encodes TAK1-binding protein 2 (TGF-β-activated kinase 1 binding protein 2). Loss of function results from truncating or, more rarely, hypomorphic intragenic variants, or from microdeletions of chromosome band 6q25.1 that remove the whole gene. The picture combines, in variable proportions, heart disease (valvular disease, frequently polyvalvular and myxomatous, congenital malformations, dilated cardiomyopathy, rhythm disturbances and vascular dilatation), short stature of prenatal or postnatal onset, generalised joint hypermobility and facial features reminiscent of Noonan syndrome [1, 2]. The National Diagnosis and Care Protocol (PNDS) published by the French Haute Autorité de Santé (HAS) in January 2026 adopts the descriptive designation «cardiac anomalies – short stature – joint hypermobility syndrome linked to variants of the TAB2 gene» and also applies to forms expressed only as heart disease [1].

1.2. Historical overview

The first interstitial deletions of the long arm of chromosome 6 including the 6q25 region were described from 1998–1999 onwards, in children with growth restriction, heart disease, facial dysmorphism and developmental delay; the region came to be treated as a 6q24-q25 contiguous gene deletion syndrome (OMIM #612863), with reports by Pirola in 1998, Kumar in 1999, Bisgaard in 2006, Caselli in 2007, Nowaczyk in 2008 and Nagamani in 2009 [7]. In 2010, Thienpont and colleagues, at the Catholic University of Leuven, compared 12 patients with 6q deletions and delineated an 850 kb critical region for congenital heart disease in 6q25.1, with five genes; bioinformatic prioritisation highlighted TAB2, whose expression was demonstrated in the human and zebrafish embryonic heart; titrated knockdown of tab2 in zebrafish produced cardiac defects from expression reductions of 41% to 58%, sequencing of 402 patients with congenital heart disease revealed two conserved missense variants, and a balanced translocation t(2;6)(q21;q25) disrupting the gene segregated in a family with aortic stenosis and arrhythmias [10, 21]. OMIM described the entity as «non-syndromic» congenital heart disease type 2 [4].

Between 2015 and 2018, reports of families with small deletions and with truncating variants multiplied, showing the myxomatous valvular component, the cardiomyopathy and the extracardiac signs: the 281 kb familial deletion of Weiss in 2015, the smallest then described to segregate with heart disease [13]; the series of Cheng in 2017, with 40 patients carrying a 6q25.1 microdeletion, which established the predisposition to cardiomyopathy with reduced systolic function even without malformation [6]; the first truncating variant (p.(Tyr497Ter)), identified by trio exome sequencing in a child with polyvalvular syndrome, by Ackerman in 2016 [22]; and the description, by Ritelli in 2018, of an Ehlers-Danlos syndrome-like phenotype with severe polyvalvular disease [23]. In 2021, Engwerda and colleagues brought together 25 of their own patients and 45 from the literature and showed that deletions and intragenic variants produce the same phenotype, proposing the name TAB2-related syndrome [2]. There followed the series of Hanson (39 carriers of point variants), Micale (who proposed the designation cardio-facio-cutaneous-articular syndrome), Westphal and Woods in 2022 [3, 12, 24, 25], and the functional assay of Xu in 2023 for the classification of missense variants [26]. OMIM changed the title to «congenital heart defects, multiple types, 2», keeping the former name in the phenotype-gene table [4]. ClinGen curation classified the association of TAB2 with this disease as definitive, with autosomal dominant inheritance, in January 2026 [27].

1.3. Nomenclature, classification and entities within scope

Orphanet distinguishes two aetiological subtypes under the same clinical picture: ORPHA:664401, the form due to a loss-of-function variant, and ORPHA:664404, the 6q25.1 microdeletion syndrome, both classified as disease subtypes and linked to OMIM phenotype #614980 [5, 16]. Large 6q24-q25 deletions including TAB2 and other genes form a continuum with the 6q25.1 microdeletion; they add manifestations attributed to neighbouring genes (more severe intrauterine growth restriction, more marked developmental delay and, in the most extensive deletions, agenesis of the corpus callosum or microcephaly) [2, 7]. A third entity, allelic but phenotypically distinct, is frontometaphyseal dysplasia associated with TAB2 missense variants, proposed by Wade in 2016 as «FMD type 3»; these variants are not loss-of-function, the patients do not have heart disease, and OMIM keeps the association as awaiting confirmation [8, 10, 24]. This article covers the haploinsufficiency syndrome as the main entity, with subsections where the two forms diverge, and describes TAB2-associated frontometaphyseal dysplasia in the differential diagnosis and molecular genetics.

2. Molecular Genetics and Pathophysiology

2.1. The gene and the protein

The TAB2 gene is located at 6q25.1 (GRCh38 coordinates 6:149,217,926–149,411,607) and encodes a protein of 693 amino acids, of about 77 kD, containing an N-terminal CUE domain (ubiquitin-binding), a region capable of forming an amphipathic coiled-coil helix (residues 535 to 609), a TAK1-binding domain and a C-terminal Npl4/RanBP2-type zinc finger (NZF) that recognises lysine 63-linked polyubiquitin chains [10, 28, 29]. The MANE Select transcript is NM_001292034.3 (protein NP_001278963.1); the clinical literature mainly uses NM_015093.5, whose coding numbering coincided with that of the MANE transcript for all the variants validated for this article [11]. In human embryos at 5.5 and 7.5 weeks, the protein is expressed in the cytoplasm of cells of the ventricular trabeculae, in the endothelium of the conotruncal cushions of the outflow tract and in the endothelium of the developing aortic valves; in zebrafish, expression becomes restricted to the cardiac outflow tract, the dorsal aorta and the posterior cardinal vein [10, 21].

2.2. The TAK1–TAB pathway and the disease mechanism

TAB2 is an adaptor that links the TAK1 kinase (encoded by MAP3K7, at 6q15) to TRAF6 and to signals from receptors such as TNFR1, the interleukin 1 receptors, Toll-like receptors and RANK. By binding Lys63 polyubiquitin chains, TAB2 promotes autophosphorylation and activation of TAK1, which activates the IKK complex (NF-κB pathway) and the MKK/JNK and p38 cascades [10, 29]. TAK1 also transmits non-canonical TGF-β and BMP signals. In 2026, Doganli and colleagues showed that TAK1 is activated in the primary cilium, through TAB2 and the catalytic α subunit of protein kinase A, during cardiomyocyte differentiation; inactivation of TAK1 inhibited ciliary signalling and differentiation, and zebrafish tak1 and tab2 mutants had cardiac and extracardiac defects [30]. The human disease results from haploinsufficiency: the remaining allele is functional, but the amount of protein is insufficient for the development of the heart and connective tissue, as demonstrated by the dose dependence in zebrafish [1, 21]. In patient fibroblasts, Micale and colleagues observed reduced NF-κB-mediated transcriptional activation, reduced TAK1 phosphorylation, impaired osteoblastic differentiation and transcriptomic dysregulation of the TGF-β, Ras-MAPK and Wnt networks, which offers an explanation for the clinical overlap with the RASopathies and connective tissue disorders [3].

Dilated cardiomyopathy has a specific mechanism, studied in models. In the mouse, cardiomyocyte-restricted deletion of Tab2 induced cardiac remodelling and heart failure; lack of TAB2 increased TNF-α-induced apoptosis and necroptosis, because TAB2, through TAK1, phosphorylates RIPK1 at serine 321 and prevents the formation of the RIPK1–FADD–caspase-8 and RIPK1–RIPK3 complexes; forced activation of TAK1 or inhibition of RIPK1 kinase activity corrected the phenotype, and genetic inactivation of RIPK1 rescued cardiac remodelling and dysfunction [10, 31]. In cardiomyocytes derived from human induced pluripotent stem cells with TAB2 knocked out, Sun and colleagues observed, at day 30 of differentiation, sarcomere disorganisation, reduced contractility, reduced energy production and mitochondrial damage; mitochondrial calcium overload appeared from day 25, owing to increased RIPK1-dependent expression of the mitochondrial calcium uniporter, and was prevented by the RIPK1 inhibitor Nec-1s [32]. The mouse with complete loss of Tab2 dies in the embryonic period from liver degeneration [10].

2.3. Variant spectrum

The great majority of the intragenic variants described are truncating (nonsense or frameshift), with some splicing variants, intragenic deletions and rare hypomorphic missense variants. In the Hanson series, 73% (11/15) were nonsense, two splicing, one frameshift and one missense; 75% (18/24) of the pathogenic variants then known lay between amino acids 35 and 535, encoded by the largest coding exon; some variants recurred in unrelated families, such as c.679C>T p.(Arg227Ter), c.1491T>A p.(Tyr497Ter), c.1636C>T p.(Arg546Ter) and c.1764+1G>A [24]. Micale and colleagues described the clustering of loss-of-function variants in a restricted linker region, which they called exon 4 according to their numbering, and a 52 kb intragenic microdeletion removing exons 1 to 3 [3]. Table 2.1 brings together illustrative variants normalised in VariantValidator.

The first missense variants described by Thienpont, NM_015093.5:c.622C>T p.(Pro208Ser) and c.688C>A p.(Gln230Lys), were associated with left ventricular outflow tract obstruction with subaortic stenosis and atrial fibrillation in a woman who died of heart failure at 61 years, and with a bicuspid aortic valve with aortic dilatation in a man; there was no family study [10, 21]. Xu and colleagues showed that, among the non-synonymous TAB2 variants, missense variants account for 295 of 576 in gnomAD and 16 of 73 in ClinVar, mostly of uncertain significance; the luciferase assay they developed measures AP-1 transcriptional activity (and not that of NF-κB, which did not discriminate), detects loss and gain of function and allowed 22 of 32 variants of uncertain significance to be reclassified; variants with partial loss of function were less associated with heart disease and more with developmental delay, hypotonia and dysmorphism, and the stability of the NZF domain proved crucial for AP-1 activation [26].

2.4. 6q25.1 microdeletions

The deletions range from 0.12 to 14.31 Mb; in the Engwerda series, the median was 8.67 Mb in the 11 patients recruited through social media and 1.76 Mb in the 25 from the literature, and the smallest region of overlap contained only TAB2 [2]. Cheng described a patient with the smallest deletion then known, limited to TAB2 [6]; the Weiss familial deletion was 281 kb (chr6:149,580,983–149,861,981, GRCh37) and included TAB2, SUMO4, ZC3H12D and PPIL4 [4, 13]. A single patient in the Engwerda series also had a deletion of CITED2, another known cardiac gene [2]. The larger deletions explain the gradation of the phenotype: moderate to severe developmental delay occurred only with deletions larger than 6.47 Mb, intrauterine growth restriction is almost exclusive to deletions and obstructive sleep apnoea has been described only with them [1, 2]. The pathogenicity of microdeletions including TAB2 is given, in a reference text on cardiogenetics, as an example of a copy number variant with an established role in dominant congenital heart disease [34].

2.5. Genotype-phenotype correlations, penetrance and expressivity

Engwerda and colleagues found no substantial phenotypic difference between deletions and intragenic variants as regards the cardiac, stature, facial and joint core [2]. The differences lie in the neurological and fetal growth signs, more frequent with deletions, especially large ones, and in hypomorphic missense variants, which are less associated with heart disease [1, 26]. Intrafamilial expressivity is highly variable: in a British family of five carriers of the same variant, the cardiac phenotypes included hypertrophic subaortic stenosis, atrial septal defect and mitral valve disease in four of the five [25]. The PNDS considers penetrance complete on the basis of cases with a detailed phenotypic description [1]; Westphal described a healthy carrier father and suggested reduced penetrance of heart disease, possibly only apparent because of subclinical disease [12]. The two readings are compatible if it is accepted that valvular and myocardial disease may appear late and go unnoticed without echocardiography.

2.6. Allelic disorders and interaction with MAP3K7

TAB2 missense variants associated with frontometaphyseal dysplasia behave differently from loss-of-function variants. The de novo variant c.1705G>A p.(Glu569Lys), located five residues upstream of the TAK1-binding interface, increased activation of ERK, p38 and JNK reporters and p38 phosphorylation, which suggests gain or alteration of function [8, 10]. Wade described two patients with FMD and TAB2 missense variants in a cohort of 20 with autosomal dominant FMD [35], and Hori reported a third, a Japanese child with de novo NM_001292034.3:c.1448C>T p.(Pro483Leu), classified as likely pathogenic, with contractures, bone deformities, clubfoot, surgically treated patent ductus arteriosus and atrial septal defect, ptosis and strabismus [15]. GeneReviews estimates that about 10% of patients with FMD have a TAB2 variant, against 50% with FLNA and 40% with MAP3K7 [9]; textbooks on bone dysplasias and clinical genetics list TAB2 as one of the genes of autosomal dominant FMD [36, 37]. The interaction is consistent with the parallel of MAP3K7, in which variants that alter TAK1 activity without abolishing it cause FMD type 2 and loss-of-function variants cause cardiospondylocarpofacial syndrome, with heart disease, hypermobility and vertebral and carpotarsal fusions [8, 38, 39].

3. Clinical Spectrum and Manifestations by System

3.1. Data sources and overall frequency of signs

The clinical data come from case reports, family series and four multicentre cohorts, all subject to ascertainment bias (patients referred for heart disease or developmental delay or recruited through social media); no estimate comes from population screening. The review carried out for the PNDS gathered 145 patients from 91 families with molecular confirmation published between 1999 and June 2025, and it is the broadest frequency base available [1]. Table 3.1 compares these frequencies with those of the main series, each with its own denominator.

Sources: [1, 2, 24, 25]. ASD, atrial septal defect; VSD, ventricular septal defect; PDA, patent ductus arteriosus; DCM, dilated cardiomyopathy; SVT, supraventricular tachycardia; FTT, failure to thrive; DDD, Deciphering Developmental Disorders study.

Only 22% (32/145) of the patients reviewed by the PNDS have the complete association of heart disease, short stature, facial features and hyperlaxity, and 15.9% (23/145) have isolated heart disease [1]. In the integrated Hanson series, 64% (25/39) had heart disease with at least one extracardiac sign and 31% (12/39) non-syndromic heart disease [24].

3.2. Cardiovascular system

Heart disease is the most frequent presenting sign and the main determinant of prognosis. Valvular disease appears at any age, more often in young people; the mitral valve is the most affected, followed by the tricuspid, aortic and pulmonary valves; the typical lesion is myxomatous dysplasia with prolapse and regurgitation, rather than stenosis, and involvement of at least two valves (polyvalvular disease), observed in about 45% of the patients reviewed, characterises the syndrome [1]. Bicuspid aortic valve (15%, 6/39, in the Hanson series), pulmonary stenosis and accessory leaflets have also been described [1, 24]. Myxomatous degeneration may be visible in the fetus: Karmegaraj described a 23-week fetus with cardiomegaly, myxomatous and redundant tricuspid and mitral leaflets, a thickened pulmonary valve and a bicuspid aortic valve, carrying a 6q25.1 deletion [40]. In adults, valvular disease progresses and may require surgery: in the Engwerda series there were mitral valve replacements at 22 and 75 years, mitral and aortic valve replacement at 51 years, aortic root repair and aortic valve replacement at 51, and aortic valve replacement at 39 years in a patient who died at 46 [2]; Permanyer described a 51-year-old man with myxomatous degeneration of all four valves and severe mitral and aortic regurgitation due to prolapse, in whose family six of nine members had valvular disease at different stages [41].

Congenital malformations were observed in 42.1% (61/145): left-to-right shunt lesions (atrial septal defect, ventricular septal defect, ductus arteriosus) in 29.7% (43/145), which manifest as high-output heart failure, with a murmur, feeding difficulties and poor weight gain; and left-sided obstructions in 7.6% (11/145), from valvular stenosis and coarctation to hypoplastic left heart syndrome; two patients had tetralogy of Fallot [1]. In the four-generation family described by Cheng, two members had hypoplastic left heart, the most recent diagnosed prenatally with short long bones and intrauterine growth restriction, and the authors proposed TAB2 deletion as a risk factor for this heart defect [42]. In the Weiss family, the proband was born with tetralogy of Fallot, short-segment pulmonary atresia, a large ventricular septal defect and myxomatous mitral and tricuspid valves, the mother had a bicuspid aortic valve and ventricular septal defect, and the grandmother underwent surgery at 22 years for myxomatous mitral regurgitation [4, 13]. This grandmother's first pregnancy ended in a stillbirth with bicuspid aortic valve, aortic stenosis and hydrops [4].

Cardiomyopathy, mainly dilated, was observed in 20% (29/145) of the patients reviewed by the PNDS, with a median age at diagnosis of 8 months (22 patients), and in 46% (18/39) of point variant carriers in the Hanson series [1, 24]. Cheng showed that cardiomyopathy with reduced systolic function can occur without structural malformation [6]. The infantile form can be severe: in the Engwerda series, two patients with a deletion died of dilated cardiomyopathy at 8 months and 4 years, and a child with a variant had dilated cardiomyopathy at 7 months, heart transplantation at 9 months and death at 2.5 years [2]; in a Finnish cohort of 66 severe childhood cardiomyopathies, with a median age at diagnosis of 0.33 years and 17 transplants, one had a TAB2 variant [43]. Cardiomyopathy also appears in adults, isolated or combined with valvular disease [24, 33]. Koene described a Dutch family in which the cardiomyopathy was detected prenatally and changed its characteristics over time [44]. Hypertrophic cardiomyopathy and left ventricular non-compaction have rarely been described [4, 24].

Rhythm and conduction disturbances (18.6%, 27/145) are associated with the malformations and the cardiomyopathy, may become apparent in the fetus and may require fetal antiarrhythmic treatment [1]. Atrial fibrillation, supraventricular tachycardia and sudden arrhythmic death have been described: in the Engwerda series, a 60-year-old man died of sudden arrhythmia and a 61-year-old woman of heart failure; in the Italian family of Ritelli, a brother with dilated cardiomyopathy and arrhythmias died suddenly at 60 years [2, 4]. Vascular dilatations (9.7%, 14/145) are mainly aortic dilatation associated with a malformed aortic valve and atrial septal aneurysm; aortic dissection has not been described [1]. In the Engwerda series, three patients from two families with a variant had a thoracic aortic aneurysm [2]. Caulfield described a 5.7 cm pulmonary artery aneurysm with severe pulmonary regurgitation [33], and Woods reported coeliac trunk stenosis, congenital stenosis of the femoral arterial tree and a subarachnoid haemorrhage with no precipitating cause [25].

3.3. Growth

Short stature affects 57.2% (83/145) of the patients reviewed, with prenatal or postnatal onset; the expected mean height is −2.3 standard deviations (SD), and may be lower, and it is sometimes disproportionate, with rhizomelic or mesomelic shortening [1]. In the Engwerda series, the median height was −2.5 SD in both own cohorts, and short stature (< −2 SD) affected 73% (8/11) of patients with a deletion and 71% (10/14) of variant carriers, in the latter mainly disproportionate [2]. Intrauterine growth restriction is almost exclusive to 6q25 deletions and does not catch up after birth [1]. Salpietro described a girl with a de novo 0.63 Mb deletion and severe prenatal and postnatal growth retardation, with an upper segment/lower segment ratio of 1.34 (above 3 SD) [45], and Deng a child with growth restriction and heart disease due to a de novo frameshift variant, NM_015093.5:c.1056del p.(Ser353fsTer8), whose product was not detected by western blot after expression in 293T cells [46]. Feeding difficulties in infancy (36%, 5/14, in the Woods series) contribute to poor weight gain [25].

3.4. Musculoskeletal system and connective tissue

Generalised joint hypermobility (42.1%) is assessed with the Beighton score; 11.5% of hyperlax patients had dislocations [1]. The other musculoskeletal anomalies (31%) include mesomelic shortening, brachydactyly, small hands and feet, flat feet, pectus excavatum and scoliosis [1]. In the Engwerda variant cohort, connective tissue abnormalities affected 93% (13/14): hypermobility in 11, flat feet in 10 and pectus excavatum in 6 [2]. Ritelli described soft, velvety and hyperextensible skin, short limbs and musculoskeletal complications of hypermobility, in a picture reminiscent of Ehlers-Danlos syndrome [23]. The skin may be soft and of abnormal texture, with marked palmar and plantar creases and easy bruising (18%), and inguinal and umbilical hernias appear more frequent than in the general population [1]. In the Koene family, the spectrum included a caudal appendage, deep sacral dimples, hypoplasia of the iliac wings and scapulae, fusion of the carpal bones, spinal canal stenosis and increased skin folds, which led to a previous clinical diagnosis of scapuloiliac dysostosis [44]. Woods also recorded delayed bone age, clinodactyly, cubitus valgus, developmental dysplasia of the hip, Perthes disease, lumbar hyperlordosis, anomalies of the C3 and C4 vertebrae and tarsal coalition [25].

3.5. Face, eyes, oral cavity and hearing

The facial features (71%) include a broad forehead, hypertelorism, short palpebral fissures slanting downwards and outwards with a hooded appearance of the eyelid, ptosis, low-set posteriorly rotated ears, a long face and midface hypoplasia; they may be subtle and suggest a RASopathy [1]. In the Engwerda deletion cohort, a broad forehead was present in all and hypertelorism, ptosis and low-set ears in 91%; 6 of 11 patients with a deletion and 7 of 14 with a variant met the van der Burgt criteria for definite Noonan syndrome [2]. The ophthalmological anomalies (31%) are, in order, ptosis (22%, 32/145), refractive errors and strabismus [1]; Woods described a patient with rod-cone dystrophy, whose relationship to the syndrome is uncertain [25]. In the oral cavity (20.7%), dental crowding or malposition and a high-arched palate each affect about 10%; three cleft palates have been described, one with Pierre Robin sequence [1], and bifid uvula [25]. In 2026, De Rosa and colleagues presented as the «first case» a TAB2-related syndrome with cleft palate [47]; the PNDS had already counted earlier cases. Hearing loss (13.8%) may be prelingual or postlingual, unilateral or bilateral, conductive, sensorineural or mixed, with few detailed descriptions [1].

3.6. Development and behaviour

Motor (31%) and language (27.6%) delay are frequently related to hypotonia and hyperlaxity; delay is more frequent with deletions, especially large ones, and the PNDS identifies no increase in the risk of intellectual developmental disorder relative to the general population [1]. The series are not homogeneous: Woods observed delay in 43% (6/14), and all patients with global delay progressed to intellectual disability, with social communication, educational and behavioural difficulties and one case of self-injurious behaviour [25]; Engwerda observed moderate to severe delay only with deletions larger than 6.47 Mb [2]; a 16-year-old Turkish boy with a 6.7 Mb deletion (6q25.1-q25.3) had hydrocephalus, cerebral palsy, autism spectrum disorder and attention deficit hyperactivity disorder [48]. Woods identified, in one patient, a second likely pathogenic variant in KMT2E, which contributed to the hypotonia, delay and microcephaly [25], a reminder that a disproportionate neurological phenotype justifies looking for another diagnosis.

3.7. Other systems

Genitourinary anomalies (7.6%) are cryptorchidism and hypospadias, described only in boys, the latter perhaps related to intrauterine growth restriction [1, 25]. Obstructive sleep apnoea (3.4%) was observed only with deletions [1]. Wandering spleen [25], hypoplastic cerebellar vermis [45] and, in a 2025 report, pancreatic malformation with heterotaxy associated with 6q25.1 haploinsufficiency [49] have also been described; in the French fetal series of Liu, a TAB2 nonsense variant was the cause of complex heart disease in a family with four affected members [50].

3.8. Allelic form: TAB2-associated frontometaphyseal dysplasia

Patients with FMD and a TAB2 missense variant have a skeletal picture very similar to that of FLNA-related FMD, with more cleft palate, scoliosis, cervical fusions, hearing loss and keloids, and the two patients reviewed by Hanson did not have heart disease [9, 24]; the Japanese child described by Hori, with the variant NM_001292034.3:c.1448C>T p.(Pro483Leu), however, had a patent ductus arteriosus and an atrial septal defect operated on at 15 months, which shows that the separation between the two allelic forms is not absolute [15]. FMD is characterised by torus-like supraorbital hyperostosis, scaphocephaly, arachnodactyly with flexion contractures and ulnar deviation, reduced joint mobility, mixed hearing loss, irregular coat-hanger ribs, flared iliac wings and metaphyseal undertubulation of the long bones, with normal height [36].

4. Clinical Recognition and Warning Signs

4.1. General principle

No manifestation is specific to the syndrome, and the diagnosis rests on the association of several signs, possibly in a family context [1]. The French PNDS organises clinical suspicion into major signs, present in more than 40% of published patients (heart disease, especially when there are several lesions; short stature of prenatal or postnatal onset; suggestive facial features; generalised joint hyperlaxity), and minor signs, present in fewer than 40% (musculoskeletal anomalies, early psychomotor developmental delay or neurodevelopmental disorder, ophthalmological anomalies, hypotonia, oral cavity anomalies, skin anomalies, deafness, inguinal or umbilical hernias and genitourinary anomalies) [1]. This list is not a set of validated criteria with known sensitivity and specificity; it is a ranking by frequency in a review of 145 published patients, subject to ascertainment bias. Because only 22% of patients have the complete association and 15.9% have isolated heart disease, suspicion should also arise in the presence of familial heart disease without extracardiac signs [1]. In the summary aimed at the attending physician, the PNDS sums up the indication to consider the diagnosis, in a child or adult, in the presence of the association of involvement of one or more valves, congenital heart malformation, dilated cardiomyopathy (especially if early) or rhythm disturbances, with short stature, joint hyperlaxity, hypotonia or orthopaedic anomalies, and facial features that may be subtle (high and broad forehead, hypertelorism, ptosis, slanting palpebral fissures, low-set posteriorly rotated ears, midface hypoplasia) [1].

4.2. Prenatal period

Prenatal involvement has been described in 30% (43/145) of patients, mainly intrauterine growth restriction and congenital heart malformation; the PNDS anticipates that the proportion will increase with improvements in fetal ultrasound and echocardiography [1]. Intrauterine growth restriction is defined by fetal biometry below the 3rd centile of the reference charts; in the syndrome it is usually proportionate (weight, length and possibly head circumference) and is more associated with 6q25 deletions (82.6%, 19/23 fetuses) than with point variants [1]. Hypoplastic left heart syndrome, tetralogy of Fallot and ventricular septal defect may be detected in the fetus; prenatal valvular disease is difficult to diagnose, but myxomatous degeneration has been documented at 23 weeks, and rhythm disturbances may appear on monitoring in late pregnancy [1, 40]. In the absence of a family history, the low specificity of these signs means that prenatal diagnosis of the syndrome has been rare (one published case in the PNDS review); increasing access to prenatal exome sequencing should change this picture [1]. In a French series of 80 fetuses with complex heart disease or heterotaxy studied with an extended panel, there was a diagnosis in 10, one of them due to a TAB2 nonsense variant [50].

4.3. Newborn and infant

At birth, congenital heart disease and hypotonia may be present, and the examination should look for inguinal or umbilical hernia, cryptorchidism or hypospadias in boys, and the facial features [1]. Left-to-right shunt lesions manifest as high-output heart failure, with a murmur, feeding difficulties and poor weight gain [1]. Dilated cardiomyopathy is diagnosed at a median of 8 months and may decompensate without prior diagnosis, which justifies intensive echocardiographic surveillance in the first years even with a normal initial echocardiogram [1, 6]. Feeding difficulties, related to hypotonia and orofacial features, affected 36% (5/14) of the patients in the Woods series [1, 25].

4.4. Child, adolescent and adult

In children, the combination of short stature (sometimes disproportionate), hyperlaxity, ptosis and a face reminiscent of Noonan syndrome, with a murmur or valvular disease, is the most suggestive picture [1, 2]. Several patients were first labelled as having Noonan syndrome, Williams syndrome, Ehlers-Danlos syndrome, Loeys-Dietz syndrome or scapuloiliac dysostosis before the molecular diagnosis [22, 25, 33, 44]. In adults, the diagnosis is suggested by myxomatous disease of several valves, familial mitral prolapse, familial dilated cardiomyopathy, atrial septal aneurysm, aortic dilatation associated with a malformed aortic valve or sudden death in the family, possibly with short stature and hypermobility [1, 28, 41]. Engwerda recommends including TAB2 in the panels used for congenital heart disease, cardiomyopathy, short stature, connective tissue disorders and suspected Noonan syndrome, as well as in rapid neonatal sequencing panels, and looking for deletions including the gene [2].

4.5. Warning and emergency signs

The PNDS asks the attending physician to recognise cardiological warning signs even in those without known heart disease [1]. Table 4.1 brings together the situations requiring rapid or urgent cardiological assessment.

Newborn screening does not include this disease, which has no biochemical marker [29].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Confirmation criterion

There are no validated clinical criteria. The PNDS defines the diagnosis as confirmed in a symptomatic individual by identification of a pathogenic (class 5) or likely pathogenic (class 4) variant responsible for TAB2 haploinsufficiency: a heterozygous deletion (copy number variant, CNV), which may include several genes, or an intragenic variant, such as small insertions or deletions or single nucleotide variants [1]. ClinGen curation regards the relationship between TAB2 and autosomal dominant congenital heart disease as definitive, which allows the ACMG/AMP classification (American College of Medical Genetics and Genomics/Association for Molecular Pathology) to be applied with the loss-of-function rule (PVS1) to null variants [27, 28]. There is no biochemical biomarker [29].

5.2. Testing strategy

The strategy recommended by the PNDS combines chromosomal microarray analysis (CGH array or SNP array) and/or exome or genome sequencing [1]. The practical order depends on the presentation, as summarised in Table 5.1.

Reference textbooks in clinical genetics recommend genomic analysis (array, panel, exome or genome) in complex heart disease or heart disease accompanied by learning difficulties or another congenital anomaly, clinical examination of both parents (with ECG and echocardiogram if there is an anomaly) and DNA sampling before cardiac surgery or extracorporeal membrane oxygenation [53]. In a series of 150 patients with syndromic heart disease studied by high-resolution CGH array, Breckpot identified 26 causal imbalances, and dysmorphism was the only predictor of a positive result [52]. The microdeletion including TAB2 is given as an example of a pathogenic CNV detectable by MLPA (multiplex ligation-dependent probe amplification), array or CNV calling from sequencing data [34].

5.3. Variant interpretation and report content

The report should state the gene, the reference transcript with version (MANE Select NM_001292034.3; the literature also uses NM_015093.5, with coinciding coding numbering for the validated variants), the HGVS (Human Genome Variation Society) coding and protein nomenclature, the genomic coordinate with the genome build, the zygosity, the ACMG/AMP class with the criteria applied and, for CNVs, the breakpoints, size and genes included [1, 11]. Null variants (nonsense, frameshift, canonical splicing, gene deletion) are the rule; missense variants are mostly of uncertain significance, and some cause partial loss of function, whereas others, rare and de novo, are associated with frontometaphyseal dysplasia through a different mechanism [8, 26]. The functional luciferase assay with an AP-1 reporter described by Xu allowed 22 of 32 variants of uncertain significance to be reclassified, but it is a research test and not a routine clinical investigation [26]. In the Chinese family of Chen, classification as pathogenic combined PVS1, PM2 (absence from population databases and from 531 controls), PP1 (cosegregation with the disease in five affected members and absence in eight unaffected members), PP3 and PP4 [28]. Family segregation requires care: valvular and myocardial disease may appear late or be subclinical, and a carrier parent with a normal echocardiogram does not exclude pathogenicity [1, 12].

5.4. Limitations of the tests

A negative result does not exclude the diagnosis. Exome or panel sequencing may fail to detect deletions of one or a few exons if there is no validated CNV calling; a low-resolution array may fail to detect small intragenic deletions, such as the 52 kb deletion removing exons 1 to 3 described by Micale [3]. Deep intronic variants, balanced rearrangements disrupting the gene (such as the t(2;6)(q21;q25) translocation described by Thienpont) and low-level mosaicism are not detected by routine methods; parental karyotype and FISH (fluorescence in situ hybridisation) are needed to identify balanced insertions, such as the paternal 6q→8q13 insertion that explained recurrence in two Brazilian siblings [10, 14, 21]. OMIM still indexes the entity, in the phenotype-gene table, as non-syndromic congenital heart disease, and some cardiomyopathy panels do not include TAB2, so it is advisable to confirm that the gene is present in the panel requested [2, 4].

5.5. Initial assessment after diagnosis

The PNDS defines the investigations to be performed at the time of diagnosis and those adapted to the manifestations (Table 5.2) [1].

When there is short stature, the general assessment described in clinical genetics textbooks includes calculation of the mid-parental target height (girls: [(father's height − 13 cm) + mother's height]/2 ± 9 cm; boys: [(mother's height + 13 cm) + father's height]/2 ± 10 cm), measurement of body segments if there is disproportion, baseline biochemistry, coeliac disease antibodies and bone age [54]. Hyperlaxity is assessed with the Beighton score (maximum 9 points), regarded as generalised with 6 or more points in prepubertal children and adolescents, 5 or more between puberty and 50 years and 4 or more after 50 years [55].

5.6. Prenatal and preimplantation diagnosis

In France, the prescription of prenatal genetic tests is decided by a multidisciplinary prenatal diagnosis committee, according to the ultrasound signs and the personal and family history; they may be chromosomal (microarray) or genic (exome or genome) [1]. When a parent is a carrier, prenatal diagnosis or preimplantation genetic testing is possible, with eligibility assessed by the multidisciplinary prenatal diagnosis centre, but the PNDS does not recommend early molecular prenatal diagnosis systematically, because the molecular result does not predict severity [1]. The conditions of access and the subsequent decisions are covered in section 8.

6. Differential Diagnosis

6.1. Comparative table

The differential diagnoses vary with the presentation: RASopathies when the face, stature and heart disease dominate; connective tissue disorders when hyperlaxity, skin and valvular disease or vasculopathy dominate; disorders of the TAK1 complex and of the filamins when there is skeletal involvement; non-syndromic familial cardiomyopathies and valvular diseases when the disease is cardiac only [1]. Table 6.1 compares the main entities.

Abbreviations: AD, autosomal dominant; AR, autosomal recessive; FMD, frontometaphyseal dysplasia.

6.2. Noonan syndrome

Noonan syndrome is the main differential diagnosis, because it shares heart disease, short stature, facial features, ptosis and early delay in similar proportions, which makes clinical distinction difficult and requires molecular investigation [1]. In the Engwerda series, 80% of patients had a face suggestive of Noonan syndrome, and more than half retrospectively met the van der Burgt criteria for definite Noonan syndrome (6 of 11 with a deletion and 7 of 14 with a variant) [2]. The same authors nevertheless consider that the «gestalt» is different, especially around the eyes, with narrow palpebral fissures, and that hypermobility and increased skin folds are not features of Noonan syndrome; in the TAB2 syndrome the mitral valve is the most affected and the lesions appear more severe, whereas in Noonan syndrome the pulmonary valve predominates [2]. In clinical genetics textbooks, dysplastic pulmonary valve stenosis affects 20% to 50% of patients with Noonan syndrome, and 7% of children with pulmonary stenosis have this syndrome [53]. Myxomatous polyvalvular disease in a patient with a Noonan phenotype and a negative RASopathy panel is the classic context for requesting TAB2 [1, 22].

6.3. Connective tissue disorders

When hyperlaxity and skin are to the fore, the first hypothesis is often Ehlers-Danlos syndrome; skin involvement is more severe in the classical type than in the TAB2 syndrome, and heart disease is not usual in Ehlers-Danlos syndrome [1]. Ritelli described in the TAB2 syndrome soft, velvety, hyperextensible skin, with severe polyvalvular disease, a picture that led to an initial diagnosis of hereditary connective tissue disorder [23]. Caulfield's patient had a Beighton score of 7 out of 9, finger and patellar dislocations since adolescence and a pulmonary artery aneurysm, and had been labelled as having Ehlers-Danlos or Loeys-Dietz syndrome [33]. Loeys-Dietz syndrome is suggested by valvular disease or vasculopathy with musculoskeletal anomalies and cleft palate, but does not usually involve growth delay, dysmorphism or developmental delay [1]. Clinical genetics textbooks point out that 5% to 10% of school-age children are hypermobile, that hypermobility is distinct from hypotonia and that the examination should look for skin changes, scars, bruising, mitral prolapse, aortic root dilatation and recurrent hernias, with an echocardiogram if there is a murmur or suspected Marfan syndrome [55].

6.4. Disorders of the TAK1 complex and of the filamins

Cardiospondylocarpofacial syndrome, caused by loss-of-function variants of MAP3K7, the molecular partner of TAB2, combines growth restriction of prenatal onset, hyperlaxity, septal defects or valvular dysplasia, brachydactyly, carpal and tarsal fusion, vertebral anomalies, conductive deafness and a suggestive face, and is the closest differential diagnosis in mechanistic terms [1, 38]. With frontometaphyseal dysplasia type 2, it forms the spectrum of MAP3K7-related disorders. The distinction is radiological (extensive carpotarsal and cervical vertebral fusions) and molecular; severe cardiomyopathy and aortic dilatation have also been described with MAP3K7, and van Woerden suggested serial ECG and echocardiographic surveillance in these patients [39, 56, 62]. Frontometaphyseal dysplasia, caused by FLNA (X-linked), by MAP3K7 or by TAB2 missense variants, is distinguished by supraorbital hyperostosis, metaphyseal undertubulation and contractures, with normal stature [9, 36]. Loss-of-function filaminopathy A may cause progressive myxomatous polyvalvular dysplasia, isolated or syndromic, but follows X-linked inheritance, and male-to-male transmission rules it out [1].

6.5. Cardiac-only forms

In families with valvular disease, dilated cardiomyopathy or sudden death without extracardiac signs, the differential diagnosis is that of non-syndromic familial cardiomyopathies and valvular diseases. In the Chen family, five carriers had heart disease only, without connective tissue anomalies [28]. The PNDS expressly includes these forms within its scope [1].

7. Treatments and Risk Management Strategies

7.1. Principles and evidence base

There is no aetiological treatment for the syndrome due to TAB2 haploinsufficiency [1]. The 2026 French PNDS, drawn up by the North-West Rare Disease Reference Centre «Developmental anomalies and malformation syndromes» (Lille University Hospital) following the 2012 HAS method, is the only dedicated guideline; it is based on a review of 44 studies identified between 2006 and 1 July 2025, of which 26 were retained (case reports, family series and descriptive cohorts), and on expert consensus, with no clinical trials [1]. The PNDS stresses that the diagnosis and treatment of each manifestation do not differ from those applied when the same manifestation occurs in isolation, and that it drew on other French PNDS (tetralogy of Fallot, univentricular heart disease, oral care of patients at risk of endocarditis, rhythm disturbances in complex adult congenital heart disease) [1]. The doses that follow therefore come from the European guidelines for the manifestations (heart failure, endocarditis) and from the official drug labels, with the extrapolation stated. The objectives defined by the PNDS are to coordinate care at a reference or competence centre, and to detect and treat early the cardiac manifestations, short stature, psychomotor and language delay, deafness, the consequences of hyperlaxity and amblyopia [1].

7.2. Organisation of care

In children, the main professionals involved are the paediatric cardiologist, the clinical geneticist and the paediatrician; in adults, the cardiologist and the general practitioner, who coordinates the other specialists (physiatrist, speech therapist, psychologist) [1]. In France, follow-up takes place in liaison with a reference or competence team of the CARDIOGEN networks (filières), for hereditary or rare heart diseases, or AnDDI-Rares, for developmental anomalies [1]. The teams involved include paediatric and adult cardiology, paediatric cardiac surgery, clinical and laboratory genetics, cytogenetics, obstetrics, physical and rehabilitation medicine, paediatric neurology, ophthalmology, otorhinolaryngology, paediatric endocrinology, dentistry and orthodontics, orthopaedics, speech therapy, psychomotor therapy, occupational therapy, adapted physical activity and psychology [1].

7.3. Cardiology

7.3.1. Surveillance of all patients

Valvular disease, dilated cardiomyopathy, rhythm disturbances and aortic dilatation may appear or progress with age. The PNDS recommends clinical examination with ECG and echocardiogram every 3 months in the first year, every 6 months until 3 years and then annually for life, justifying the initial frequency by the incidence of early dilated cardiomyopathy [1]. Section 4.6 of the PNDS indicates an annual cardiological assessment in childhood, which corresponds to the phase after 3 years of the same schedule [1]. Engwerda and Cheng also recommend regular surveillance for cardiomyopathy and thoracic aneurysm, even with a normal initial echocardiogram, because valvular disease, aneurysm and cardiomyopathy may appear late [2, 6].

7.3.2. Congenital heart malformations

The treatment of the malformations (atrial and ventricular septal defects, ductus arteriosus, coarctation, tetralogy of Fallot, hypoplastic left heart) does not differ from usual practice [1]. The general approach to congenital heart disease includes, for a small ventricular septal defect, surveillance and prevention of endocarditis and, for a moderate or large defect, diuretics and angiotensin-converting enzyme (ACE) inhibitors with surgical closure if there is volume overload [63]. Coexisting cardiomyopathy modifies the surgical risk: Ugaki described two Japanese boys with a ventricular septal defect and a dilated ventricle with reduced function; the first, aged 7 months, with Klinefelter syndrome and a TAB2 variant, had severe heart failure and pulmonary hypertension and was operated on with patch closure, nitric oxide and inotropes, extubated on day 7 and in NYHA (New York Heart Association) class I at 2 years; the second, aged 16 months, with a 6q24.3-q25.1 deletion and cardiomyopathy treated with carvedilol and enalapril since 3 months, had postoperative biventricular failure with extracorporeal membrane oxygenation for 7 days and Pseudomonas mediastinitis, was extubated on day 45 and was in class I at 15 months of follow-up [64]. The authors concluded that surgery can be performed, but that the postoperative course of these patients requires preparation for ventricular dysfunction [64].

7.3.3. Valvular and polyvalvular disease

The decision on severe valvular disease with haemodynamic impact, or on polyvalvular disease, lies with the cardiology and cardiac surgery team, weighing the characteristics of the lesion, the possibility of a repair that preserves the native valve versus replacement, and comorbidities; follow-up is adapted to the lesion and the procedure [1]. Published operations include mitral valve replacements at 22 and 75 years, mitral and aortic valve replacement at 51, aortic root repair and aortic valve replacement at 51 and aortic valve replacement at 39 years, and, in Caulfield's patient, resection of a pulmonary artery aneurysm with implantation of a 27 mm pulmonary valve homograft at 35 years, with improvement of the ejection fraction from 22% to 45%; the pulmonary valve showed myxoid degeneration [2, 33]. In the combined Engwerda series, 13 of 70 patients (19%) underwent cardiac surgery [2]. The myxomatous and multivalvular nature of the disease should be taken into account in the choice between repair and replacement, because other valves may later require intervention [1, 41].

7.3.4. Dilated cardiomyopathy

The PNDS refers to the French expert consensus on hereditary heart diseases (February 2018): dilated cardiomyopathy does not require treatment different from that of other causes of heart failure; pharmacological treatment of heart failure with reduced ejection fraction and assessment of arrhythmic risk, including implantation of a cardioverter-defibrillator for primary prevention, are discussed, comorbidities and other cardiovascular risk factors are treated, and follow-up takes place every 6 to 12 months, with clinical examination, ECG and echocardiogram [1]. Several patients had a favourable course with heart failure treatment, which could later be stopped; conversely, there were early deaths from undiagnosed decompensated cardiomyopathy and a transplant at 9 months [1, 2]. The 2021 European Society of Cardiology (ESC) guideline for adults rests on four drug classes: ACE inhibitor or angiotensin receptor-neprilysin inhibitor (ARNI), beta-blocker, mineralocorticoid receptor antagonist and sodium-glucose cotransporter 2 (SGLT2) inhibitor [65]. Table 7.1 brings together the ESC adult doses, alongside the US label where read.

The doses in Table 7.1 are adult doses. In children, the use of these drugs in heart failure is off-label and extrapolated; the US carvedilol label reports a double-blind trial in 161 children (mean age 6 years; 2 months to 17 years; 45% under 2 years) with chronic heart failure, with no significant effect on the composite clinical outcome at 8 months, and with chest pain (17% versus 6%), dizziness (13% versus 2%) and dyspnoea (11% versus 0%) more frequent than with placebo [68]. Ugaki's child received carvedilol and enalapril from 3 months, but the report does not state the doses [64]. The US labels describe symptomatic hypotension after the first dose of enalapril in patients on a diuretic, which leads to stopping the diuretic two to three days before starting the ACE inhibitor when possible, and, during carvedilol titration, dizziness and fluid retention, managed by increasing the diuretic without mandatory discontinuation, and bradycardia, which requires dose reduction [66, 68].

7.3.5. Rhythm disturbances

The antiarrhythmic drug is chosen according to the type and mechanism of the disturbance and the cardiac comorbidities, and the PNDS refers to the French PNDS on rhythm and conduction disturbances in complex adult congenital heart disease [1]. Arrhythmias detected in the fetus may require fetal pharmacological treatment [1]. Sudden arrhythmic death has been described in adults and in a 2-year-old child with biventricular dilatation, which justifies assessment of arrhythmic risk in families [2, 28].

7.3.6. Aortic dilatation

Aortic dilatation requires regular surveillance of the diameter and its rate of change. Owing to the lack of data on its course and because aortic dissection has not been described, the PNDS does not recommend preventive beta-blockade systematically; the decision is individual, and treatment of arterial hypertension forms part of prevention [1]. This position differs from practice in syndromic connective tissue aortopathies, in which dissection is a documented risk. The three thoracic aortic aneurysms in two families of the Engwerda series and Caulfield's pulmonary artery aneurysm show that surveillance should include the ascending aorta and the pulmonary artery [2, 33].

7.3.7. Prevention of infective endocarditis

The PNDS classifies patients as at high risk or intermediate risk of infective endocarditis according to the heart disease [1]. High-risk patients are those with previous endocarditis; with a mechanical or biological prosthetic valve, whatever the technique, or prosthetic material used for valve repair; with cyanotic congenital heart disease, with prosthetic material such as valved conduits, systemic-to-pulmonary shunts or atrial or ventricular septal defect closure devices (up to 6 months after the procedure); and with ventricular assist devices [1]. Intermediate-risk patients are those with a morphological anomaly or dysfunction (regurgitation or stenosis) of one or more of the four valves, with hypertrophic obstructive cardiomyopathy, with high-risk congenital heart disease and with intracardiac pacing devices [1]. Because valvular disease affects 68.3% of patients, most belong at least to the intermediate category [1]. General measures apply to all: perfect oral hygiene, strict skin hygiene, disinfection of wounds, antibiotic treatment of focal bacterial infections and discouragement of antibiotic self-medication [1]. Antibiotic prophylaxis is considered according to the risk and the dental procedure (extraction, manipulation of the gingiva or periapical region); in intermediate risk, the decision is discussed with the cardiologist [1]. The 2023 ESC guideline is concordant: it recommends prophylaxis in high-risk patients undergoing at-risk dental procedures and regards patients with degenerative valve disease or congenital valve anomalies, including bicuspid aortic valve, as at intermediate risk, without routine prophylaxis; in this guideline, isolated congenital valve anomalies do not count as high-risk congenital heart disease [51]. Table 7.2 gives the 2023 ESC doses.

Cephalosporins are not used after anaphylaxis, angioedema or urticaria with penicillin, and clindamycin is no longer recommended because of the risk of Clostridioides difficile infection [51].

7.4. Growth and nutrition

7.4.1. Surveillance and investigation

Weight and height are measured every 3 months until 2 years and then every 6 months, on general population charts; in the presence of deceleration of growth velocity, height below −2 SD or below −1.5 SD of the target height, the PNDS recommends a paediatric endocrinology consultation, and assessment of pubertal development looks for early or delayed puberty [1]. Investigation of short stature includes measurement of body segments (arm span and sitting height), an endocrine work-up (TSH, free T4, ACTH, 8 a.m. cortisol, IGF-1, IGFBP3 and, depending on pubertal stage, FSH, LH and testosterone or oestrogens), a general work-up (full blood count, electrolytes, renal and liver function, total IgA and anti-transglutaminase antibodies, testing for proteinuria) and bone age; if growth delay is confirmed, a growth hormone stimulation test is performed to rule out deficiency [1]. Intrauterine growth restriction requires monitoring of catch-up growth; the PNDS states that 95% of children born with growth restriction catch up and 5% remain short without treatment, but the reference data are less favourable: in the Swedish Gothenburg cohort (3650 term newborns followed to about 18 years), 13.4% of children born with a length below −2 SD were still below −2 SD at 1 year and 7.9% in adulthood, and the 2007 international consensus estimates that about 10% of children born small for gestational age fail to catch up [1, 69, 70]. In the syndrome, the growth restriction associated with deletions does not usually catch up [1]. Oral feeding disorders with undernutrition, associated with hypotonia and orofacial dysmorphism, justify gastroenterological and speech therapy assessment, including of swallowing [1].

7.4.2. Growth hormone

Few published patients have received growth hormone (somatropin), and its efficacy has not been established [1]. Engwerda mentions a patient from the literature, with a 6q24.2-q25.2 deletion and growth hormone deficiency, treated with good results, and a patient in their cohort eligible for treatment [2]. In France, growth hormone is indicated in growth hormone deficiency and, without deficiency, in situations such as the child born small for gestational age (≤ −2 SD) without catch-up at 4 years and with height ≤ −3 SD, a frequent situation in this syndrome; the PNDS recommends assessing efficacy after 1 year and stopping if there is no benefit, with clinical monitoring at least every six months and annual laboratory monitoring (IGF-1, fasting blood glucose, free T4, glycated haemoglobin) and education in self-injection [1]. The PNDS does not state doses; Table 7.3 gives the label doses for the «small for gestational age» indication, which is the applicable analogue, and those for growth hormone deficiency.

The European dose of 0.035 mg/kg/day corresponds to about 0.245 mg/kg/week, half the US maximum [71, 73]. The contraindications in the SmPC include active tumour, closed epiphyses and acute critical illness with complications after open-heart surgery, abdominal surgery, multiple trauma or acute respiratory failure; in two trials with 522 adults in these conditions, mortality was 42% with 5.3 or 8 mg/day of somatropin versus 19% with placebo [71]. This contraindication is directly relevant in a syndrome in which cardiac surgery is frequent: treatment should not be given during a complicated postoperative course after cardiac surgery [71]. The SmPC requires monitoring for scoliosis, which may progress with rapid growth, examination of the hip in a child who limps (slipped capital femoral epiphysis), looking for papilloedema in the presence of severe headache, visual disturbance or vomiting (benign intracranial hypertension) and considering pancreatitis in the presence of abdominal pain [71]. Scoliosis is already part of the syndrome, which reinforces this surveillance [1]. There are no published data on the effect of growth hormone on the cardiomyopathy or the valves of these patients.

7.5. Orthopaedics, physical medicine and hyperlaxity

Annual orthopaedic follow-up is sufficient, with attention to spinal alignment at the onset of puberty [1]. Pain associated with hyperlaxity, with recurrent dislocations of large and small joints, is treated with analgesics adapted to its intensity and frequency; depending on the joint, physiotherapy (proprioceptive work, muscle strengthening), compression garments and orthoses are used, and in children a team with pain medicine, physical and rehabilitation medicine, psychomotor therapy and occupational therapy may be needed [1]. Scoliosis is treated first with physiotherapy (flexibility and postural correction); a progressive curve greater than 20° may require a night-time or daytime brace, and excessive progression leads to surgical discussion with the paediatric orthopaedic surgeon [1]. Asymptomatic bilateral flat feet that correct on tiptoe walking do not need treatment; symptomatic flat feet are seen in an orthopaedic clinic and treated, in most cases, with insoles [1]. Pectus excavatum is corrected if there is functional impact, possibly for cosmetic reasons, with chest computed tomography and preoperative pulmonary function tests [1].

7.6. Development and schooling

The developmental trajectory is not well characterised and the profiles are diverse, so support is individualised: physiotherapy and psychomotor therapy if there is hypotonia in the first years, occupational therapy if the impact hampers learning to write, and speech therapy for delay in oral or written language [1]. In France, a child under 4 years with delay may be referred to an early medical-social action centre (CAMSP) until 6 years, or, if no place is available, to a coordination and referral platform that funds private-practice support; psychometric assessment is desirable when there is delay in acquisitions, to define strengths and weaknesses and adapt schooling [1]. Mainstream schooling is the first option when there is no intellectual developmental disorder, with a support assistant for pupils with disabilities if needed; a mild disorder may justify localised units for school inclusion, and major cognitive difficulties a medical-educational institute, with regular reassessment [1].

7.7. Eyes, teeth and hearing

General population eye screening (first week of life, 9 months, 2.5 years and 6 years) detects refractive errors and strabismus; in the presence of signs, the examination is performed promptly at any age, because the risk is amblyopia, which can lead to monocular blindness [1]. Ptosis, present in about 20% of patients, is corrected surgically when it obstructs the visual axis [1]. Oral hygiene is important, especially in patients at risk of endocarditis, with consultations at least every six months; malpositions and malocclusions may require orthodontics, and patients with an intellectual developmental disorder may benefit from the French Handident network [1]. Early detection of deafness is essential because of its impact on language and behaviour; treatment is the usual otorhinolaryngological treatment, with effective treatment of infections to avoid otitis media with effusion and, if necessary, grommets, tonsillectomy, adenoidectomy or hearing aids [1].

7.8. Therapeutic education and associations

There is no specific therapeutic education programme; in patients with valvular disease, heart failure or arrhythmia, education focuses on warning signs, triggers of acute events, adapted physical activity and monitoring of blood pressure and heart rhythm on treatment; in those treated with growth hormone, on self-injection [1]. Referral to patient associations should always be offered; in France there is no specific association, but there are rare disease associations [1]. Disclosure of the diagnosis, psychological support and social rights (including the French long-term condition scheme for conditions not on the list) are described in section 8.

7.9. Pregnancy, delivery and the perioperative period

A woman with a pathogenic variant should have a preconception cardiology consultation, to assess whether there is a contraindication to pregnancy, adapt treatment and anticipate complications; during pregnancy, cardiological follow-up is regular if there is heart disease [1]. Regurgitant valve lesions are generally well tolerated, whereas mitral or tricuspid stenosis may worsen with the increase in cardiac output in the 2nd trimester and maternal tachycardia [1]. Vaginal delivery is generally possible with stable heart disease; caesarean section may be recommended for untreated tight mitral or tricuspid stenosis or for severe, poorly tolerated heart disease [1]. The preconception review includes the cardiovascular drugs being taken, in line with the general guidance for women with congenital heart disease [1, 53]. The sources consulted do not describe specific anaesthetic recommendations; perioperatively, ventricular function and the valves should be assessed before any surgery, in the light of the postoperative dysfunction described by Ugaki, and endocarditis prophylaxis follows Table 7.2 when the procedure warrants it [51, 64].

7.10. Transition to adulthood

Annual cardiological follow-up continues for life, and the assessment comes to focus on the functional and socio-occupational impact of hyperlaxity [1]. Transition should ensure that the adult cardiologist is aware of the risk of valvular progression, cardiomyopathy and aneurysm in adulthood, and that reproductive counselling is repeated [1, 2].

7.11. Agents and practices to avoid

Antibiotic self-medication should be discouraged [1]. Clindamycin is not used for endocarditis prophylaxis, and cephalosporins are not used after a severe immediate reaction to penicillin [51]. Somatropin is contraindicated in acute critical illness after open-heart surgery [71]. According to the PNDS, beta-blockers are not used routinely in isolated aortic dilatation [1].

7.12. Regulatory status

No medicine is authorised for the syndrome; the drugs used are those for the manifestations. Table 7.4 summarises the status of the main ones.

7.13. Reference studies and limitations of the evidence

There are no registered clinical trials for the syndrome; a ClinicalTrials.gov search for the terms TAB2 and 6q25 returned only irrelevant studies [74]. The evidence on natural history and response to treatment comes from the descriptive series in Table 7.5.

None of these studies compares treatments. The frequencies come from populations ascertained through disease, biased towards severe cases and heart disease, and with short follow-up in children; the lower median age of the Engwerda own cohorts (4 years for deletions and 11.5 years for variants, against 9.5 and 30 years in the literature) may have underestimated late-onset cardiomyopathy [2]. The carvedilol and somatropin trials did not include patients with this syndrome, and their results cannot be transposed without reservation [68, 72].

7.14. New developments and research

In the 12 months before November 2026, the new developments were observational or preclinical. At the regulatory and guideline level, the publication of the PNDS by the HAS on 26 January 2026 and the ClinGen classification as definitive in January 2026 [1, 27]. At the clinical level, reports of ventricular septal defect repair in children with cardiomyopathy (Ugaki, 2026), of polyvalvular dysplasia in Indian children with normal ventricular function and a Beighton score of 7 out of 9 (Krishna, 2025), of cleft palate presented as a first description (De Rosa, 2026) and of autism and attention deficit in a 6q25.1-q25.3 deletion (Sezgin, 2026), the latter with stabilisation on olanzapine 5 mg/day, a report that does not constitute a recommendation [47, 48, 64, 75]. At the experimental level, Doganli showed, in association studies (an approximately twofold higher burden of rare protein-altering variants in TAK1, TAB2 and PRKACA in 1471 patients with syndromic heart disease, compared with 2405 with non-syndromic heart disease and 45,082 controls) and in zebrafish, that TAK1 acts in the primary cilium in non-canonical TGF-β/BMP signalling during cardiac development [30]. Sun showed in human stem cell-derived cardiomyocytes that RIPK1-dependent mitochondrial calcium overload contributes to the cardiomyopathy and that the RIPK1 inhibitor Nec-1s prevents it in vitro [32]; together with the work of Yin in the mouse, these data point to RIPK1 as a hypothetical therapeutic target for dilated cardiomyopathy due to TAB2 deficiency, without any study in humans [31, 32].

There is no gene therapy or targeted treatment in clinical development for this disease. In a haploinsufficiency disorder, the theoretically applicable strategies would be upregulation of the normal allele (by targeted transcriptional activation or by oligonucleotides acting on RNA regulatory elements) or gene replacement by viral vector in the myocardium; none has been tested for TAB2 in the sources consulted, not even in animal models, and the strict dose dependence demonstrated in zebrafish, in which reductions of 41% to 58% in expression are enough to cause cardiac defects, would imply that excess expression would also have to be assessed as a risk, as suggested by the gain-of-function variants associated with frontometaphyseal dysplasia [8, 10, 21]. The intervention under study closest to the clinic is pharmacological and preclinical (RIPK1 inhibition), and no targeted therapy has gone beyond the preclinical stage [31, 32].

7.15. Divergences between sources

The sources diverge on four points. First, on the cardiological schedule: the summary and section 5.3.1 of the PNDS give a three-monthly frequency in the 1st year and six-monthly until 3 years, whereas section 4.6 refers to an annual assessment in childhood; the more intensive schedule is the one in the summary and in the follow-up table of the same PNDS, and is that of Table 11.1 [1]. Second, on aortic dilatation: the PNDS does not recommend systematic beta-blockade, unlike practice in syndromic connective tissue aortopathies [1]. Third, on growth hormone: the PNDS applies the French criterion (≤ −3 SD), the European SmPC requires < −2.5 SD at 4 years and the US label allows treatment from 2 years, with a maximum weekly dose about twice as high [1, 71, 73]. Fourth, on fetal surveillance: the PNDS proposes specialised fetal echocardiography from 18 weeks of gestation and a follow-up scan in the 3rd trimester when a parent is a carrier, whereas the general guidance for complex familial heart disease in a clinical genetics textbook suggests echocardiograms at 14, 19 and 22 weeks [1, 53]. Penetrance is also described as complete by the PNDS and as possibly reduced by Westphal (section 2.5) [1, 12]. On endocarditis prevention, the PNDS and the 2023 ESC guideline agree on risk stratification, and the PNDS leaves prophylaxis in intermediate risk to discussion with the cardiologist [1, 51]. On catch-up growth after intrauterine growth restriction, the 95% proportion given by the PNDS is higher than that of the cohort data and the international consensus (about 90%, with 7.9% short stature in adulthood in a Swedish cohort), which serve as the reference (section 7.4.1) [1, 69, 70].

8. Reproductive Options and Family Counselling

8.1. Disclosure of the diagnosis and support

The PNDS recommends that disclosure take place in a suitable setting and at a dedicated time, if possible with both parents, possibly involving the person who requested the molecular test, a physician familiar with the disease and, depending on the context, a psychologist or a nurse, drawing on the HAS guide for the disclosure of chronic disease [1]. Before disclosure, it is advisable to know the personal and family history, the religious and cultural beliefs and the socioeconomic level, which may make the news more traumatic; during disclosure, the characteristics of the syndrome and the surveillance arrangements are explained, with emphasis on lifelong cardiological follow-up [1]. A post-disclosure consultation allows the diagnosis to be reviewed, questions that have arisen to be answered and transmission to be explained; a written report should be given to the patient and to the professionals, and psychological support should always be offered [1]. In France, 100% coverage may be requested as a long-term condition not on the list, together with a certificate for the Maison Départementale des Personnes Handicapées (MDPH), the departmental body that recognises disability and the associated rights [1].

8.2. Recurrence risks

The disease is autosomal dominant, with variable expressivity and, according to the PNDS, apparently complete penetrance in cases with a detailed phenotypic description [1]. Table 8.1 summarises the risks by scenario.

The empirical recurrence risks used in isolated congenital heart disease (for example, about 1.4% for the same defect in a sibling, according to a Danish cohort of more than 18,000 patients) do not apply when an identified Mendelian cause exists, in which case counselling follows the mode of inheritance [53].

8.3. Informing the family and cascade testing

In a familial form, informing at-risk relatives is mandatory, because preventive measures exist (screening for cardiac anomalies, prevention of paroxysmal rhythm disturbances); this may be done by the patient or by the geneticist, at the patient's request, using the contact details the patient provides [1]. Testing targeted at the familial variant may be offered to minors and adults at risk, because the heart disease is progressive and dilated cardiomyopathy or valvular disease require early care and adapted follow-up [1]. This indication in minors is justified by direct medical benefit, unlike adult-onset diseases without preventive measures. In the Chen family, testing of 13 relatives identified five carriers, all with some echocardiographic abnormality, and eight non-carriers without heart disease, which made it possible to discontinue surveillance of the latter [28]. The absence of heart disease in an adult relative does not exclude carrier status [12].

8.4. Prenatal and preimplantation diagnosis

When a member of the couple may transmit the disease, prenatal diagnosis or preimplantation genetic testing is possible, with eligibility assessed by the multidisciplinary prenatal diagnosis centre [1]. The PNDS does not recommend early molecular prenatal diagnosis systematically; it may be discussed case by case, only at the parents' request, after information on the limits of prognostic assessment [1]. The usual prenatal follow-up, which detects growth restriction and morphological anomalies, is supplemented by specialised fetal echocardiography from 18 weeks of gestation, by a follow-up echocardiogram in the 3rd trimester and by monitoring of the fetal heart rhythm [1]. Decisions on termination of pregnancy for medical reasons rest mainly on the presence of severe heart disease documented by ultrasound and discussed collegially [1]. When a de novo variant is discovered after an ultrasound finding, the prognosis depends on the severity of the heart disease and not on the molecular result alone; without severe cardiac malformation, the other manifestations are generally compatible with a favourable course with adequate postnatal care [1]. In the Chen family, amniocentesis at 19 weeks identified the familial variant in a fetus with a normal echocardiogram at 15 weeks; the report does not describe the outcome of the pregnancy [28].

8.5. Content of the counselling consultation

The consultation should explain the 50% risk per pregnancy for a carrier, the unpredictability of severity, the possibility of late-onset heart disease, the risk of about 1% in de novo cases, the indication to study the parents (including echocardiography, even if asymptomatic) and the reproductive options available [1, 53]. For deletions, the microarray report should make it possible to say whether the deletion includes other genes, which modifies expectations regarding development and fetal growth [1, 2].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence

Prevalence is unknown; Orphanet does not publish an estimate, and the PNDS counted 145 patients from 91 families with molecular confirmation published between 1999 and June 2025 [1, 16]. This number is a count of published cases and not a prevalence. The frequency of TAB2 in congenital heart disease cohorts gives an order of magnitude of the gene's contribution: two missense variants in 402 patients with congenital heart disease in the original Thienpont series, one nonsense variant in 80 fetuses with complex heart disease in the French series of Liu and one variant in 66 children with severe cardiomyopathy in the Finnish cohort of Vasilescu [21, 43, 50]. Doganli observed an approximately twofold higher burden of rare variants in TAB2 and TAK1 in 1471 patients with syndromic heart disease than in controls [30]. In reference textbooks, structural congenital heart disease affects 7 to 8 per 1000 live births, which places the TAB2 syndrome as a rare cause among many [53, 63].

9.2. Geographical distribution and populations studied

The reports come mainly from Europe and North America: Belgium (the founding Leuven series), the Netherlands (Engwerda; Koene), the United Kingdom (Woods; Salpietro), Italy (Ritelli; Micale), Germany (Westphal), Spain (Permanyer; Sufrate-Vergara), France (Liu; PNDS) and Finland (Vasilescu); the United States (Cheng, Hanson, Weiss, Ackerman, Caulfield, Xu); and, in Asia, China (Chen, Deng), Japan (Hori, Ugaki) and India (Krishna) [2, 3, 6, 12, 13, 15, 21–26, 28, 33, 41, 43–46, 49, 50, 64, 75]. No founder variants or clustering in isolated populations have been described; some variants, such as p.(Arg546Ter) and p.(Tyr497Ter), recurred in unrelated families, which indicates independent recurrent mutation and not a demonstrated founder effect [24]. In Brazil, the team at the Federal University of São Paulo described two siblings with an 8.5 Mb interstitial deletion at 6q24.2-q25.2, inherited from a balanced paternal insertion into 8q13 detected by FISH, with 10 years of follow-up: developmental delay, microcephaly, short stature, dysmorphism, multiple anomalies and intellectual disability, and stenosis of the descending aorta in one of them [14]. The sources consulted do not include cases from Portugal. Data by ethnicity and ancestry are insufficient to compare expression between populations.

9.3. Special populations

In pregnancy, the risk depends on the valvular lesion (stenoses worsen in the 2nd trimester) and on the cardiomyopathy, as described in section 7.9 [1]. In patients with large deletions and intellectual disability, dental care may require specialised networks and adapted communication of pain and cardiac symptoms [1]. In adolescents and adults with hypermobility and valvular disease, prescription of adapted physical activity forms part of therapeutic education [1]. The coexistence of another genetic disorder (Klinefelter syndrome in one of Ugaki's patients; a KMT2E variant in one of Woods's patients) has been described and should be considered when the phenotype is atypical [25, 64].

9.4. European reference networks and patient associations

The relevant European reference networks are ERN ITHACA (intellectual disability, telehealth, autism and congenital anomalies), for malformation syndromes, ERN GUARD-Heart (rare and low-prevalence heart diseases), for heart disease and cardiomyopathy, and ERN BOND (rare bone diseases), for the allelic form of frontometaphyseal dysplasia [18–20]. In France, patients are followed in the AnDDI-Rares networks (with the reference and competence centres for developmental anomalies and malformation syndromes) and CARDIOGEN (centres for cardiomyopathies, rhythm disturbances and hereditary or rare heart diseases), the coordinating centre of the PNDS being the North-West Reference Centre for Developmental Anomalies and Malformation Syndromes, at the Jeanne de Flandre Hospital of Lille University Hospital; the Fondation Maladies Rares and Orphanet are indicated as resources [1]. There is no specific association in France; the Chromosome 6 Project, an international social media-based project, brings together families with chromosome 6 alterations and made it possible to recruit the Engwerda deletion cohort [1, 2, 76].

10. Illustrative Clinical Case

In 2020, Chen and colleagues, from the Jiangxi Maternal and Child Health Hospital (Nanchang, China), described a three-generation family with 25 members (11 men, 12 women and two fetuses) [28]. The proband (II:5), aged 31 years, was first seen at 15 weeks of pregnancy; she had had mild tachypnoea and occasional fatigue since the age of 25, and a transthoracic echocardiogram at 30 years had shown mild aortic stenosis with mild mitral, tricuspid and aortic regurgitation; she had never undergone surgery or received medication [28]. Her first daughter (III:7) had dilatation of both ventricles and died suddenly at 2 years; the second pregnancy (III:8) was terminated after absence of fetal heartbeat at 9 weeks; the third daughter (III:9), aged 2 years, had no symptoms or echocardiographic abnormalities [28]. The proband's mother (I:2) died of sudden arrhythmia at the end of the fifth decade. Three sisters had atrial septal aneurysm; one (II:1, diagnosed at 39 years) also had dilatation of the left coronary artery, confirmed by coronary angiography, and mild aortic regurgitation; another (II:7) had mild mitral prolapse with mitral and pulmonary regurgitation; the third (II:9) had dilatation of the left atrium and left ventricle; a 3-year-old nephew (III:11) had mild dilatation of the left ventricle and right atrium and mitral prolapse with regurgitation [28]. No relative had connective tissue anomalies or other extracardiac signs described [1, 28].

Exome sequencing of the proband (mean depth 186.76×, alignment to hg19) identified the heterozygous variant NM_015093.5:c.446C>G p.(Ser149Ter), which corresponds, on the MANE transcript, to NM_001292034.3:c.446C>G p.(Ser149Ter) (NC_000006.12:g.149378361C>G) [11, 28]. The variant creates a premature stop codon that eliminates most of the protein, including the TAK1-binding domain; it was confirmed by Sanger sequencing in the five living affected individuals (II:1, II:5, II:7, II:9 and III:11), was absent in the eight unaffected individuals tested and in 531 controls, and was classified as pathogenic (PVS1, PM2, PP1, PP3 and PP4) [28]. The figures of the article, consulted through the legends in the full text, show the pedigree with the chromatogram of the variant (Figure 1, whose legend mistakenly refers to nucleotide 466), the diagram of the TAB2 domains with the N-terminal CUE domain and the C-terminal TAK1-binding domain (Figure 2) and the ACMG/AMP criteria applied (Figure 3) [28]. The fetal echocardiogram at 15 weeks was normal; with the couple's consent, amniocentesis at 19 weeks identified the variant in the fetus (III:10), regarded as being at high risk of heart disease; the article does not describe the outcome of the pregnancy [28].

The case shows the cardiac-only form of the syndrome, the intrafamilial expressivity ranging from mild lesions in adults to sudden death in childhood, and the value of cascade testing to separate the relatives who require surveillance. Two management decisions in the report differ from the 2026 PNDS: molecular prenatal diagnosis was recommended to the proband, whereas the PNDS does not recommend it routinely and accepts it only at the parents' request after information on the prognostic limits; and surveillance of the carrier fetus is not described, whereas the PNDS recommends specialised fetal echocardiography from 18 weeks, a follow-up scan in the 3rd trimester and monitoring of the fetal rhythm [1, 28]. According to the PNDS, the proband, with several valves affected, would have an indication for regular cardiological follow-up during pregnancy, and the paediatric carriers for echocardiogram and ECG every 3 months in the 1st year and every 6 months until 3 years [1].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

The vital prognosis depends mainly on the severity and course of the heart disease [1]. In the neonatal period, the determinants are complex malformations, dilated cardiomyopathy and rhythm disturbances; the PNDS counts four early deaths in the literature, due to complex malformations or undiagnosed decompensated dilated cardiomyopathy, and several patients with a good course after treatment of heart failure, which could be stopped [1]. Outside these situations, the course after adequate treatment is mostly favourable, but requires lifelong cardiological follow-up [1]. In the combined Engwerda series (n = 70), 19% of patients underwent cardiac surgery, four died of late effects of structural heart disease and three of dilated cardiomyopathy in childhood; the published late outcomes include valve operations between 22 and 75 years, a patient on the transplant waiting list, sudden arrhythmic death at 60 years and death from heart failure at 61 [2]. These figures come from populations with severity bias and short follow-up in children; they are not survival rates. Final height, development and autonomy vary with the type of alteration: large deletions are associated with more marked delay and growth restriction without catch-up, and the PNDS identifies no increase in the risk of intellectual developmental disorder in the variant forms [1, 2]. There are no data on life expectancy or quality of life.

11.2. Surveillance schedule

Table 11.1 brings together the PNDS schedule and the targeted investigations.

For the dental examination of patients at risk of endocarditis, the PNDS follow-up table indicates an «annual or twice-yearly» frequency, and the stomatology section specifies that follow-up should be at least twice a year, which corresponds to the six-monthly interval in Table 11.1 [1].

12. Summary and Final Considerations

TAB2-related syndrome results from haploinsufficiency of the gene that links the TAK1 kinase to cytokine and TGF-β signalling, through loss-of-function variants or 6q25.1 microdeletions, and is an autosomal dominant disease of highly variable expressivity. Its core is cardiac: myxomatous valvular disease, often of several valves and with onset at any age, congenital malformations, dilated cardiomyopathy that may appear in infancy, rhythm disturbances and aortic dilatation; in smaller proportions, it adds short stature, joint hypermobility and a face reminiscent of Noonan syndrome, and, in extensive deletions, intrauterine growth restriction and more marked developmental delay. Only one fifth of published patients have the complete picture, and about one in six have heart disease only, so the diagnosis depends on including TAB2 and CNV analysis in the tests requested for heart disease, cardiomyopathy, short stature, connective tissue disorders and suspected Noonan syndrome with a negative panel. The 2026 French PNDS is the first dedicated guideline and rests on descriptive series and consensus: surveillance by echocardiogram and ECG every 3 months in the 1st year, every 6 months until 3 years and then annually for life; usual treatment of each lesion; no systematic beta-blockade in aortic dilatation; endocarditis prophylaxis according to risk; growth hormone possible under the small for gestational age criteria, with unproven efficacy; rehabilitation for hyperlaxity; and specialised fetal surveillance in families. Counselling rests on a 50% risk per pregnancy for carriers, about 1% in de novo cases and higher when there is a balanced parental rearrangement, and on cascade testing, which is also justified in minors. The main gaps are the absence of prevalence and penetrance estimates from population screening, the short follow-up of children, the lack of data on growth hormone and the absence of targeted treatment, despite preclinical data pointing to the RIPK1 pathway in the cardiomyopathy.

13. References

1. Centre de Référence Maladies Rares « Anomalies du développement et syndromes malformatifs » Nord-Ouest, CHU de Lille. Protocole National de Diagnostic et de Soins (PNDS) : Syndrome d’anomalies cardiaques – petite taille – hypermobilité articulaire lié aux variants du gène TAB2. Texte du PNDS, argumentaire et synthèse à destination du médecin traitant. Saint-Denis La Plaine: Haute Autorité de Santé; décembre 2025 (publié le 26 janvier 2026). Disponible sur: https://www.has-sante.fr/jcms/p_3826258
2. Engwerda A, Leenders EKSM, Frentz B, Terhal PA, Löhner K, de Vries BBA, et al. TAB2 deletions and variants cause a highly recognisable syndrome with mitral valve disease, cardiomyopathy, short stature and hypermobility. Eur J Hum Genet. 2021;29(11):1669-76. doi:10.1038/s41431-021-00948-0
3. Micale L, Morlino S, Carbone A, Carissimo A, Nardella G, Fusco C, et al. Loss-of-function variants in exon 4 of TAB2 cause a recognizable multisystem disorder with cardiovascular, facial, cutaneous, and musculoskeletal involvement. Genet Med. 2022;24(2):439-53. doi:10.1016/j.gim.2021.10.009
4. Online Mendelian Inheritance in Man, OMIM®. #614980 Congenital heart defects, multiple types, 2; CHTD2. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/614980
5. Orphanet. 6q25.1 microdeletion syndrome (ORPHA:664404). Available from: https://www.orpha.net/en/disease/detail/664404
6. Cheng A, Dinulos MBP, Neufeld-Kaiser W, Rosenfeld J, Kyriss M, Madan-Khetarpal S, et al. 6q25.1 (TAB2) microdeletion syndrome: congenital heart defects and cardiomyopathy. Am J Med Genet A. 2017;173(7):1848-57. doi:10.1002/ajmg.a.38254
7. Online Mendelian Inheritance in Man, OMIM®. #612863 Chromosome 6q24-q25 deletion syndrome. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/612863
8. Wade EM, Daniel PB, Jenkins ZA, McInerney-Leo A, Leo P, Morgan T, et al. Mutations in MAP3K7 that alter the activity of the TAK1 signaling complex cause frontometaphyseal dysplasia. Am J Hum Genet. 2016;99(2):392-406. doi:10.1016/j.ajhg.2016.05.024
9. Robertson S, Wade E. FLNA-Related Otopalatodigital Spectrum Disorders. 2005 Nov 30 [updated 2025 Jun 26]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1393/
10. Online Mendelian Inheritance in Man, OMIM®. *605101 TAK1-binding protein 2; TAB2. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/605101
11. VariantValidator. Validation of TAB2 variants on GRCh38 (NM_001292034.3, MANE Select). Available from: https://variantvalidator.org
12. Westphal DS, Mastantuono E, Seidel H, Riedhammer KM, Hahn A, Vill K, et al. There is more to it than just congenital heart defects - The phenotypic spectrum of TAB2-related syndrome. Gene. 2022;814:146167. doi:10.1016/j.gene.2021.146167
13. Weiss K, Applegate C, Wang T, Batista DA. Familial TAB2 microdeletion and congenital heart defects including unusual valve dysplasia and tetralogy of fallot. Am J Med Genet A. 2015;167A(11):2702-6. doi:10.1002/ajmg.a.37210
14. Meloni VA, Guilherme RS, Oliveira MM, Migliavacca M, Takeno SS, Sobreira NL, et al. Cytogenomic delineation and clinical follow-up of two siblings with an 8.5 Mb 6q24.2-q25.2 deletion inherited from a paternal insertion. Am J Med Genet A. 2014;164A(9):2378-84. doi:10.1002/ajmg.a.36631
15. Hori A, Migita O, Kawaguchi-Kawata R, Narumi-Kishimoto Y, Takada F, Hata K. A novel TAB2 mutation detected in a putative case of frontometaphyseal dysplasia. Hum Genome Var. 2021;8(1):40. doi:10.1038/s41439-021-00166-6
16. Orphanet. Cardiac anomalies-short stature-joint hypermobility-facial dysmorphism syndrome due to TAB2 mutation (ORPHA:664401). Available from: https://www.orpha.net/en/disease/detail/664401
17. Orphanet. Frontometaphyseal dysplasia (ORPHA:1826). Available from: https://www.orpha.net/en/disease/detail/1826
18. European Reference Network ITHACA (Intellectual disability, TeleHealth, Autism and Congenital Anomalies). Available from: https://ern-ithaca.eu
19. European Reference Network GUARD-Heart (rare and low prevalence complex diseases of the heart). Available from: https://guardheart.ern-net.eu
20. European Reference Network on Rare Bone Diseases (ERN BOND). Available from: https://ernbond.eu
21. Thienpont B, Zhang L, Postma AV, Breckpot J, Tranchevent LC, Van Loo P, et al. Haploinsufficiency of TAB2 causes congenital heart defects in humans. Am J Hum Genet. 2010;86(6):839-49. doi:10.1016/j.ajhg.2010.04.011
22. Ackerman JP, Smestad JA, Tester DJ, Qureshi MY, Crabb BA, Mendelsohn NJ, et al. Whole exome sequencing, familial genomic triangulation, and systems biology converge to identify a novel nonsense mutation in TAB2-encoded TGF-beta activated kinase 1 in a child with polyvalvular syndrome. Congenit Heart Dis. 2016;11(5):452-61. doi:10.1111/chd.12400
23. Ritelli M, Morlino S, Giacopuzzi E, Bernardini L, Torres B, Santoro G, et al. A recognizable systemic connective tissue disorder with polyvalvular heart dystrophy and dysmorphism associated with TAB2 mutations. Clin Genet. 2018;93(1):126-33. doi:10.1111/cge.13032
24. Hanson J, Brezavar D, Hughes S, Amudhavalli S, Fleming E, Zhou D, et al. TAB2 variants cause cardiovascular heart disease, connective tissue disorder, and developmental delay. Clin Genet. 2022;101(2):214-20. doi:10.1111/cge.14085
25. Woods E, Marson I, Coci E, Spiller M, Kumar A, Brady A, et al. Expanding the phenotype of TAB2 variants and literature review. Am J Med Genet A. 2022;188(11):3331-42. doi:10.1002/ajmg.a.62949
26. Xu W, Graves A, Weisz-Hubshman M, Hegazy L, Magyar C, Liu Z, et al. Assigning pathogenicity for TAB2 variants using a novel scalable functional assay and expanding TAB2 disease spectrum. Hum Mol Genet. 2023;32(6):959-70. doi:10.1093/hmg/ddac252
27. ClinGen. Gene-disease validity: TAB2 – congenital heart defects, multiple types, 2 (Definitive, autosomal dominant; 2026-01-05). Available from: https://search.clinicalgenome.org/kb/gene-validity/CGGV:assertion_d7a67505-817d-4405-b2c3-d6d06c1e16f7-2026-01-05T170000.000Z
28. Chen J, Yuan H, Xie K, Wang X, Tan L, Zou Y, et al. A novel TAB2 nonsense mutation (p.S149X) causing autosomal dominant congenital heart defects: a case report of a Chinese family. BMC Cardiovasc Disord. 2020;20(1):27. doi:10.1186/s12872-019-01322-1
29. GeneGuidance. TAB2 gene record (dump 2026-09-26).
30. Doganli C, Thomsen OK, Baird DA, Ali Y, Sarusie MVK, Audain E, et al. TAK1 operates at the primary cilium in non-canonical TGFB/BMP signaling to control heart development. PLoS Biol. 2026;24(8):e3003902. doi:10.1371/journal.pbio.3003902
31. Yin H, Guo X, Chen Y, Zeng Y, Mo X, Hong S, et al. TAB2 deficiency induces dilated cardiomyopathy by promoting RIPK1-dependent apoptosis and necroptosis. J Clin Invest. 2022;132(4):e152297. doi:10.1172/JCI152297
32. Sun W, Zhang J, Li S, Fu W, Liu Y, Liu M, et al. TAB2 deficiency induces dilated cardiomyopathy by promoting mitochondrial calcium overload in human iPSC-derived cardiomyocytes. Mol Med. 2025;31(1):42. doi:10.1186/s10020-025-01103-x
33. Caulfield TR, Richter JE, Brown EE, Mohammad AN, Judge DP, Atwal PS. Protein molecular modeling techniques investigating novel TAB2 variant R347X causing cardiomyopathy and congenital heart defects in multigenerational family. Mol Genet Genomic Med. 2018;6(4):666-72. doi:10.1002/mgg3.401
34. Jongbloed JDH, Lekanne Deprez RH, Vatta M. Introduction to Molecular Genetics. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 3–26.
35. Wade EM, Jenkins ZA, Daniel PB, Morgan T, Addor MC, Adès LC, et al. Autosomal dominant frontometaphyseal dysplasia: delineation of the clinical phenotype. Am J Med Genet A. 2017;173(7):1739-46. doi:10.1002/ajmg.a.38267
36. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 326–330.
37. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 198–199.
38. Le Goff C, Rogers C, Le Goff W, Pinto G, Bonnet D, Chrabieh M, et al. Heterozygous mutations in MAP3K7, encoding TGF-β-activated kinase 1, cause cardiospondylocarpofacial syndrome. Am J Hum Genet. 2016;99(2):407-13. doi:10.1016/j.ajhg.2016.06.005
39. van Woerden GM, Senden R, de Konink C, Trezza RA, Baban A, Bassetti JA, et al. The MAP3K7 gene: further delineation of clinical characteristics and genotype/phenotype correlations. Hum Mutat. 2022;43(10):1377-95. doi:10.1002/humu.24425
40. Karmegaraj B. Myxomatous degeneration of cardiac valves in a fetus with 6q25.1 (TAB2) deletion. Cardiol Young. 2024;34(2):459-61. doi:10.1017/S1047951123004419
41. Permanyer E, Laurie S, Blasco-Lucas A, Maldonado G, Amador-Catalan A, Ferrer-Curriu G, et al. A single nucleotide deletion resulting in a frameshift in exon 4 of TAB2 is associated with a polyvalular syndrome. Eur J Med Genet. 2020;63(4):103854. doi:10.1016/j.ejmg.2020.103854
42. Cheng A, Neufeld-Kaiser W, Byers PH, Liu YJ. 6q25.1 (TAB2) microdeletion is a risk factor for hypoplastic left heart: a case report that expands the phenotype. BMC Cardiovasc Disord. 2020;20(1):137. doi:10.1186/s12872-020-01404-5
43. Vasilescu C, Ojala TH, Brilhante V, Ojanen S, Hinterding HM, Palin E, et al. Genetic basis of severe childhood-onset cardiomyopathies. J Am Coll Cardiol. 2018;72(19):2324-38. doi:10.1016/j.jacc.2018.08.2171
44. Koene S, Klerx-Melis F, Roest AAW, Kleijwegt MC, Bootsma M, Haak MC, et al. Sacral abnormalities including caudal appendage, skeletal dysplasia, and prenatal cardiomyopathy associated with a pathogenic TAB2 variant in a 3-generation family. Am J Med Genet A. 2022;188(12):3510-5. doi:10.1002/ajmg.a.62929
45. Salpietro V, Ruggieri M, Mankad K, Di Rosa G, Granata F, Loddo I, et al. A de novo 0.63 Mb 6q25.1 deletion associated with growth failure, congenital heart defect, underdeveloped cerebellar vermis, abnormal cutaneous elasticity and joint laxity. Am J Med Genet A. 2015;167A(9):2042-51. doi:10.1002/ajmg.a.37118
46. Deng Q, Wang X, Gao J, Xia X, Wang Y, Zhang Y, et al. Growth restriction and congenital heart disease caused by a novel TAB2 mutation: a case report. Exp Ther Med. 2023;25(6):258. doi:10.3892/etm.2023.11957
47. De Rosa A, Kalantari S, Carboni M, Casella A, Giorgio E, Apicella A, et al. Expanding the phenotype of TAB2-related syndrome: the first case with cleft palate and insights into palatal development. Am J Med Genet A. 2026;200(6):1427-30. doi:10.1002/ajmg.a.70092
48. Sezgin M, Koyuncu O, Güneş M, Yıldırım Budak B. Comorbid autism spectrum disorder and attention-deficit/hyperactivity disorder in a patient with 6q25.1-q25.3 microdeletion: a case report. Dev Neuropsychol. 2026;51(3-4):207-14. doi:10.1080/87565641.2026.2674063
49. Sufrate-Vergara B, Martín Grau C, Pedrola Vidal L, Marín-Reina P, Esteban Ríos MJ, Rodríguez-López R. Pancreatic malformation and heterotaxy syndrome associated with 6q25.1 haploinsufficiency. Pancreas. 2025;54(6):e571-2. doi:10.1097/MPA.0000000000002473
50. Liu H, Giguet-Valard AG, Simonet T, Szenker-Ravi E, Lambert L, Vincent-Delorme C, et al. Next-generation sequencing in a series of 80 fetuses with complex cardiac malformations and/or heterotaxy. Hum Mutat. 2020;41(12):2167-78. doi:10.1002/humu.24132
51. Delgado V, Ajmone Marsan N, de Waha S, Bonaros N, Brida M, Burri H, et al. 2023 ESC Guidelines for the management of endocarditis. Eur Heart J. 2023;44(39):3948-4042. doi:10.1093/eurheartj/ehad193
52. Breckpot J, Thienpont B, Peeters H, de Ravel T, Singer A, Rayyan M, et al. Array comparative genomic hybridization as a diagnostic tool for syndromic heart defects. J Pediatr. 2010;156(5):810-7. doi:10.1016/j.jpeds.2009.11.049
53. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 112–115.
54. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 312–315.
55. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 176–177.
56. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 344–345.
57. Online Mendelian Inheritance in Man, OMIM®. #157800 Cardiospondylocarpofacial syndrome; CSCF. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/157800
58. Online Mendelian Inheritance in Man, OMIM®. *602614 Mitogen-activated protein kinase kinase kinase 7; MAP3K7. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/602614
59. Orphanet. Cardiospondylocarpofacial syndrome (ORPHA:3238). Available from: https://www.orpha.net/en/disease/detail/3238
60. Online Mendelian Inheritance in Man, OMIM®. #617137 Frontometaphyseal dysplasia 2; FMD2. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/617137
61. Online Mendelian Inheritance in Man, OMIM®. #305620 Frontometaphyseal dysplasia 1; FMD1. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/305620
62. Morlino S, Castori M, Dordoni C, Cinquina V, Santoro G, Grammatico P, et al. A novel MAP3K7 splice mutation causes cardiospondylocarpofacial syndrome with features of hereditary connective tissue disorder. Eur J Hum Genet. 2018;26(4):582-6. doi:10.1038/s41431-017-0079-x
63. Holanda RP, Ribeiro EM. Cardiopatia congênita. In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 155–159.
64. Ugaki S, Ogura S, Tsuno K, Shimizu T, Masuda S, Mafune R, et al. Repair of ventricular septal defect in children with TAB2 gene anomalies and associated cardiomyopathy. Surg Case Rep. 2026;12(1):26-33. doi:10.70352/scrj.cr.26-0033
65. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021;42(36):3599-726. doi:10.1093/eurheartj/ehab368
66. VASOTEC (enalapril maleate) tablets. Prescribing information. NDA 018998. Revised 2026 Jan. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:VASOTEC
67. Enalapril maleate oral solution. Prescribing information. Revised 2024 Apr. Available from: https://api.fda.gov/drug/label.json?search=openfda.generic_name:ENALAPRIL
68. COREG (carvedilol) tablets. Prescribing information. NDA 020297. Revised 2026 Jul. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:COREG
69. Karlberg J, Albertsson-Wikland K. Growth in full-term small-for-gestational-age infants: from birth to final height. Pediatr Res. 1995;38(5):733-9. doi:10.1203/00006450-199511000-00017. PMID: 8552442.
70. Clayton PE, Cianfarani S, Czernichow P, Johannsson G, Rapaport R, Rogol A. Management of the child born small for gestational age through to adulthood: a consensus statement of the International Societies of Pediatric Endocrinology and the Growth Hormone Research Society. J Clin Endocrinol Metab. 2007;92(3):804-10. doi:10.1210/jc.2006-2017. PMID: 17200164.
71. European Medicines Agency. Omnitrope (somatropin): EPAR – Product information (summary of product characteristics). First authorised 12 April 2006. Available from: https://www.ema.europa.eu/en/documents/product-information/omnitrope-epar-product-information_en.pdf
72. INFARMED. Genotropin 5,3 mg/1 ml e 12 mg/1 ml pó e solvente para solução injetável (somatropina): resumo das características do medicamento. Titular: Pfizer. Lisboa: INFARMED, I.P. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=3847&tipo_doc=rcm
73. Pfizer. GENOTROPIN (somatropin) for injection. Prescribing information. BLA 020280. Revised 2026 Aug. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:GENOTROPIN
74. ClinicalTrials.gov. Search for the terms «TAB2» and «6q25» [database]. Bethesda (MD): National Library of Medicine; accessed 2 November 2026. Available from: https://clinicaltrials.gov
75. Krishna MR, Sennaiyan UN. Polyvalvular dysplasia related to haploinsufficiency of the TAB2 gene in Indian children - a disease with need for careful follow-up. Ann Pediatr Cardiol. 2025;18(3):298-300. doi:10.4103/apc.apc_85_25
76. Chromosome 6 Project. Available from: https://www.chromosome6.org