Feingold Syndrome

Class of malformation syndromes caused by reduced dosage of the MYCN–miR-17~92 axis: Feingold syndrome type 1 (MYCN), Feingold syndrome type 2 (MIR17HG deletion at 13q31.3), contiguous gene deletions at 2p24 and, as a mirror image, megalencephaly-polydactyly syndrome due to MYCN gain of function.

Literature and regulatory information updated: 26 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Founding descriptions

The first description is attributed to Murray Feingold, who in 1975 published a familial case in a syndrome identification journal; the report has no PubMed record and is cited through GeneReviews, OMIM and Cognet et al. [3, 26–28]. OMIM summarises it as a father, a son and a grandmother with microcephaly, hand anomalies, tracheoesophageal fistula, duodenal atresia and normal intelligence, with father-to-son transmission; according to the reconstruction by Klaniewska et al., the proband was a boy with type C tracheoesophageal fistula, duodenal atresia, microcephaly, micrognathia, apparent absence of the middle phalanx of the 5th finger and 2–3 toe syndactyly, and the digital anomalies and microcephaly of the father and paternal grandmother occurred with normal intelligence [27, 29]. The version by Samara et al., which describes the family differently, diverges from these two sources [15]. In 1978, Feingold described a mother and daughter with similar findings, without fistula or atresia, and in 1990 König et al. named the condition, in a mother and son, microcephaly, mesobrachyphalangy and tracheoesophageal fistula syndrome (MMT) [27]. The same eponym also independently designates Floating-Harbor syndrome, described by Pelletier and Feingold in 1973, so the two conditions must not be confused [30].

In 1991, Brunner and Winter, from Nijmegen, described two families with autosomal dominant transmission of hand and foot anomalies, short palpebral fissures and variable microcephaly with learning difficulties; between one third and one quarter of those affected were born with oesophageal atresia, duodenal atresia or both, some isolated patients had hypoplastic thumbs or congenital heart disease, and the phenotype resembled that of 13q22-qter deletions, although the karyotype was normal [31]. This resemblance to deletions of the long arm of chromosome 13 anticipated the identification of type 2 by two decades. In 1997, Courtens et al. described the seventh family, in which the proband had oesophageal and duodenal atresia and the mother, an aunt, an uncle and the maternal grandmother had the same phenotype; they estimated that about one third of patients had intestinal atresia or obstruction and recorded the synonyms then in use [32]. In the same year, Kawame, Pagon and Hudgins described four patients with microcephaly, normal intelligence and digital anomalies, whom they considered a distinct entity; in 2006 Teszas et al. identified the MYCN variant p.(Glu73Ter) in a 4-year-old boy with the classic form, whose carrier mother and grandmother had microcephaly and digital anomalies with normal intelligence, and proposed that this picture was a mild form of the syndrome [27, 33]. Also in 1997, Feingold et al. reviewed 6 new families (12 patients), with learning difficulties or intellectual disability in 87%, transmission through at least two generations in 8 of 11 families and intrafamilial variability mainly in the gastrointestinal findings; Innis et al., pooling the published cases, estimated a risk of tracheoesophageal fistula and/or duodenal atresia of 8 in 29 (about 28%); and Frydman et al., who used the designation MODED, found microcephaly in 78%, oesophageal or duodenal atresia in 25% (16.6% after correction for ascertainment bias) and learning difficulties in 31% [27].

1.2. Mapping and gene identification

In 2000, Celli et al. studied four pedigrees under the designation oculodigitoesophagoduodenal syndrome, including a three-generation Dutch family with 11 affected individuals; after excluding 12 genomic regions, they obtained linkage to marker D2S390 with a maximum LOD score of 4.51 (θ = 0), and a submicroscopic deletion in a fourth family narrowed the critical region to 7.3 cM at 2p23-p24 [34]. In 2003, the same group's clinical review counted at least 79 patients in 25 families, with complete penetrance and variable expressivity [35]. In 2005, van Bokhoven et al. found in one family a microdeletion of at most 1.2 Mb containing only MYCN and, in a cohort of 23 families, 12 different heterozygous variants in 15 families, 6 of them de novo, most of them truncating near or within the regions encoding the bHLH and leucine zipper domains, and 3 missense variants at two adjacent arginine residues of the bHLH domain [27, 36]. All the variants were predicted to affect both the full-length protein and the short ΔMYCN isoform, and the authors concluded that early embryogenesis and postnatal brain growth depend on MYCN dosage [37]. In 2008, Marcelis et al. analysed 93 patients from 50 families with strong clinical suspicion, found 16 variants in 17 families (26 patients) and pooled the 77 patients with a MYCN variant or deletion described at the time, from whom they established the genotype-phenotype correlations and feature frequencies that are still used as a reference [27, 38].

Genetic heterogeneity became apparent when Cognet et al. found a MYCN variant or deletion in only 8 of 17 French patients with a clinical diagnosis (47%), with no alterations in the conserved regulatory elements of the locus [28]. In 2011, de Pontual et al. used array comparative genomic hybridisation (aCGH) to study ten index cases with the skeletal features of the syndrome and no MYCN variant, and found in two of them germline hemizygous 13q31.3 microdeletions removing the polycistronic miR-17~92 cluster; mice with targeted deletion of the cluster reproduced the cardinal features, and the work was the first example of a microRNA (miRNA) gene responsible for a human developmental syndrome [13]. One of the index cases in that series (AO39) belonged to the group of patients without a MYCN variant in the Cognet series, with a microcephalic father and sister who also had digital anomalies [13, 28]. In 2013, Tassano et al. described a de novo interstitial 13q31.1–q32.1 deletion including the cluster, in a report that already used the designation "Feingold syndrome 2" in its title [39].

The third milestone was the description of the opposite phenotype. In 2019, Kato et al. identified by trio exome sequencing the de novo MYCN variant c.173C>T p.(Thr58Met) in a Japanese patient with megalencephaly, ventriculomegaly, hypoplastic corpus callosum, intellectual disability, polydactyly and neuroblastoma [40]. In 2023, Nishio et al. added two patients, a French patient with p.(Pro60Leu) and a Japanese patient with p.(Thr58Met), demonstrated stabilisation of the protein, reproduced the phenotype in a mouse model, and named the condition megalencephaly-polydactyly syndrome, a "mirror phenotype" of Feingold syndrome [12].

1.3. Current nomenclature and classifications

Orphanet classifies Feingold syndrome (ORPHA:1305) as a rare congenital malformation syndrome with two subtypes, distinguished by the presence (type 1, ORPHA:391641) or absence (type 2, ORPHA:391646) of gastrointestinal atresia, both with code Q87.8 of the International Classification of Diseases, 10th revision (ICD-10) [1, 2, 4]. In OMIM, type 1 corresponds to 164280 and type 2 to 614326; MYCN gain of function has entry 620748 [3, 19]. One curation database assigns type 1 the code C69.2 and the category of neoplastic disease, a mapping error inconsistent with Orphanet, whose code is Q87.8 [2, 9]. ClinGen (Clinical Genome Resource) classified the MYCN–Feingold syndrome type 1 relationship as definitive on 2 July 2025 and the MYCN–megalencephaly-polydactyly syndrome relationship as moderate on 7 January 2026; MIR17HG has no validity or dosage curation [7]. In 2011, de Pontual et al. left open whether type 2 was a true Feingold syndrome or a distinct syndromic brachydactyly, and supported the link between the two types by showing that MYC and MYCN activate transcription of miR-17~92; Orphanet now treats the two conditions as subtypes of the same entity [1, 13].

1.4. Entities in scope and overview of the class

Feingold syndrome type 1 (FS1) is defined by the association of brachymesophalangy of the 2nd and 5th fingers, toe syndactyly, microcephaly, short palpebral fissures, gastrointestinal atresias (mainly oesophageal and duodenal) and mild to moderate learning difficulty, caused by MYCN haploinsufficiency [2, 3]. Feingold syndrome type 2 (FS2) shares microcephaly, short stature, brachymesophalangy, toe syndactyly and thumb hypoplasia, without gastrointestinal atresia, and results from hemizygous deletion of MIR17HG [3, 4]. Contiguous gene deletions at 2p24 that include MYCN produce the type 1 phenotype with additional features and are grouped with type 1 [5]. Megalencephaly-polydactyly syndrome (OMIM 620748, "Megalencephaly-polydactyly syndrome", official acronym MPAPA; called "megalencephaly-polydactyly syndrome" in ClinGen and "Megalencephaly-postaxial polydactyly-ventriculomegaly-neuroblastoma syndrome" in Orphanet, ORPHA:730286), caused by MYCN gain of function, is the allelic disorder with the opposite mechanism [6–8, 12]. Two mirror-image copy-number variants complete the axis: the germline 2p24 duplication including MYCN and DDX1, with predisposition to neuroblastoma and Wilms tumour, and the 13q31.3 microduplication including MIR17HG, with macrocephaly and tall stature [23, 41]. Somatic MYCN amplification in tumours is an acquired event with no germline implication [3, 19].

AD, autosomal dominant; CMA, chromosomal microarray; CNV, copy-number variant.

2. Molecular Genetics and Pathophysiology

2.1. The MYCN gene and the N-Myc protein

MYCN belongs to the MYC family, together with c-Myc and L-Myc, and was identified by homology with c-Myc in amplified sequences from neuroblastomas [3]. In neuroblastoma, somatic MYCN amplification is the most relevant genomic alteration and is associated with a poor prognosis, whereas hereditary predisposition is due to germline variants in ALK and PHOX2B [43]; MYCN amplification also explains a small percentage of non-hereditary retinoblastomas with normal RB1 [44]. These are somatic alterations, distinct from the germline mechanisms of the class. The gene has three exons; although exon 1 contains a potential initiation codon, protein synthesis begins at the first ATG of exon 2 through an internal ribosome entry site (IRES) located in the 5' untranslated region, with increased activity in neuronal cells [3, 45]. The canonical protein is encoded by exons 2 and 3, with an N-terminal transactivation domain and, in the C-terminal region, the basic helix-loop-helix (bHLH) and leucine zipper domains, required for dimerisation with another bHLH partner and for DNA binding [9, 46]. There is also a short isoform, ΔMYCN, which is likewise affected by the pathogenic variants of the syndrome [37, 38]. The MANE Select transcript (Matched Annotation from NCBI and EMBL-EBI) is NM_005378.6, which encodes NP_005369.2; version NM_005378.4, used in GeneReviews, encodes the same protein with the same c. numbering [10].

N-Myc is a nuclear transcription factor that regulates progenitor proliferation and differentiation during development [9]. The deduced protein has 456 amino acids and about 49 kD, with the greatest similarity to MYC in the C-terminal DNA-binding domain; transcription starts from two promoters, with two alternative first exons (1a and 1b) that join exon 2, and the GC-rich promoter is bidirectional, also driving antisense transcription of MYCNOS (NCYM) [36]. In the second-trimester human fetus, MYCN RNA is abundant in the cerebral germinal layer, the primordial cortex and the inner nuclear and ganglion layers of the primitive retina, and disappears with differentiation [36]. In the human embryo, Cognet et al. observed ubiquitous expression at Carnegie stage 13, most intense in the limb bud mesenchyme; at stage 15, high expression in the central and peripheral nervous system, the oesophageal and bronchial epithelia, Rathke's pouch, the sympathetic ganglia and the ectodermal and mesenchymal components of the upper limb; and, at stages 17–18, in the developing pituitary, the adrenal gland and the hindgut, with no cardiac expression at stage 18 [28]. Expression in Rathke's pouch led the authors to hypothesise a pituitary contribution to the growth deficit [28]. In the mouse, Mycn is highly expressed at E13.5 and declines to a minimum at P15 [19].

2.2. MYCN haploinsufficiency and tissue pathophysiology

The mechanism of type 1 is haploinsufficiency: all germline variants described reduce the amount of functional protein, whether through gene deletion, a premature termination codon (PTC) or substitution of residues in the DNA-binding domain [3, 38]. Mouse models explain the distribution of malformations. Mycn-null mice die between E10.5 and E12.5 with developmental delay and small mesonephros, lung, heart and intestine, and mice with 25% of normal protein levels die at birth from severely deficient lung branching [19, 28]. N-Myc is required in neurogenesis for the rapid expansion of progenitor populations [47], and in the developing limb it links control of skeletal size to digit separation [48]; conditional knockout in skeletal mesenchymal stem cells reproduces the limb anomalies of the syndrome through reduced PI3K (phosphatidylinositol 3-kinase) signalling [12]. In the murine cerebellum, Nmyc is a target of Sonic Hedgehog signalling and sustains proliferation of granule neuron progenitors, and in the retina it regulates progenitor proliferation through repression of p27(Kip1) [36]. A hypomorphic model has asplenia, a finding that Cognet et al. recorded in a patient with a MYCN variant who also had a diaphragmatic hernia [28]. The regulatory regions of Mycn have been characterised in transgenic mice [49], but Cognet et al. found no variants in five conserved non-coding elements and no microrearrangements in a 3.3 Mb region of the locus in patients without a coding variant [28].

2.3. MYCN variant spectrum in type 1

The spectrum includes whole-gene and contiguous gene deletions, intragenic deletions and duplications, and missense, nonsense, frameshift and splice variants [3]. In the compilation by Marcelis et al., the 77 patients with a molecular diagnosis had 4 missense variants in the DNA-binding domain, 19 variants with a PTC (six subject to nonsense-mediated decay, NMD) and 5 larger deletions [38]. Truncating variants are distributed across exons 2 and 3; pathogenic missense variants cluster in exon 3, in the bHLH domain, at residues such as Arg382, Arg391, Arg393, Arg394 and Pro409 [46]. The variant c.64C>T p.(Gln22Ter) in exon 1, which affects only the ΔMYCN isoform initiated in that exon, was found by Marcelis et al. in a proband who also had p.(Glu73Ter); p.(Glu73Ter) segregated with the disease in the mother and maternal grandmother, whereas p.(Gln22Ter) was in the unaffected father, indicating that variants restricted to ΔMYCN do not cause the syndrome [3, 36]. The effect of variants on other alternative transcripts remains unknown [3]. OMIM lists nine selected allelic variants of MYCN.

In October 2026, ClinVar contained 74 pathogenic, 45 likely pathogenic and 209 uncertain significance records in MYCN [50].

The source notations were normalised in VariantValidator to the MANE transcript and checked against ClinVar; no conversion was made by hand [10, 50].

The variant c.454G>A p.(Ala152Thr), described in 2026 in a family from Málaga, illustrates the limits of interpreting missense variants in the N-terminal domain: it co-segregated in two informative meioses and was almost absent from gnomAD v4.1.0 (1 in 1,431,958 alleles), but in silico predictions pointed to benignity (REVEL 0.018; AlphaMissense 0.074) and conservation was low to moderate, so it was classified as a variant of uncertain significance under the 2024 Bayesian framework of the Association for Clinical Genomic Science (ACGS) [46].

2.4. Contiguous gene deletions at 2p24

Deletions encompassing MYCN and adjacent genes were considered relatively common by GeneReviews, which counted nine individuals in 2019, most identified by CMA during investigation of multiple congenital anomalies [3]. Chen et al. described a 4.4 Mb de novo deletion at 2p24.3→p24.2 including FAM84A, NBAS, DDX1, MYCNOS and MYCN in a girl with profound bilateral hearing loss and inner ear malformation [21]. Burnside et al. pooled six individuals with overlapping deletions of 3.4 to 16.8 Mb and defined a smallest region of overlap of 1.53 Mb (chr2:14,614,477–16,148,021, hg19) with five genes: NBAS, DDX1, MYCNUT, MYCNOS and MYCN; type 1 features dominated the picture [5]. According to the synthesis by Samara et al., larger deletions are associated with more severe intellectual disability and, of the genes in the region, only SOX11 and MYCN have a known haploinsufficiency effect [15]. De novo deletions and variants of SOX11 cause on their own a neurodevelopmental disorder with features of Coffin-Siris syndrome [42]. The smallest deletions described are 425 kb, with MYCN alone [28], and about 342 kb, with DDX1, MYCNUT, MYCNOS and MYCN [15].

2.5. MIR17HG, the miR-17~92 cluster and type 2

MIR17HG does not encode a protein: it is the host gene of a polycistronic transcript with six miRNAs (miR-17, miR-18a, miR-19a, miR-19b-1, miR-20a and miR-92a-1) that modulate survival, proliferation, differentiation and angiogenesis [11, 17]. MYC and MYCN directly activate its transcription [13]. In deletion carriers, expression of the six miRNAs in leucocytes is about 50% of normal, with no allelic compensation [13]. Mice heterozygous for the targeted deletion (miR-17~92Δ/+) are viable, fertile and smaller, with shortening of the middle phalanx of the 5th digit, a smaller skull shortened in the anteroposterior axis and no syndactyly, and the deletion does not alter Gpc5 expression in the limbs, which attributes the features to Mir17-92 and not to Gpc5 [53]; homozygotes die in the perinatal period with severely delayed endochondral and membranous ossification, absence of the middle phalanx of the 5th digit, hypoplasia of the 1st ray, fusion of cervical vertebrae, microcephaly and fusion of the proximal carpal bones [13]. The same heterozygous mice are more active in the open field and, as adults, show selective deficits in spatial ability, social novelty recognition and memory, with dopamine alterations in the medial prefrontal cortex and serotonin alterations in the hippocampus [53, 54].

All pathogenic alleles described are deletions, from 165 kb to 17.2 Mb [11, 14]. The 165 kb deletion in family AO70 removed only the cluster and the first exon of GPC5, and the GPC5 deletions found in healthy controls indicate that isolated loss of this gene does not explain the phenotype [13]. Muriello et al. point out that the region covered by the larger deletions contains 141 additional genes, 31 with an OMIM entry and 7 with an associated phenotype, and that, of the 18 DECIPHER cases with a MIR17HG deletion, five had some type 2 features [14]. Neighbouring genes explain features that do not belong to the core of the syndrome: keratoconus in a patient with a 17.2 Mb deletion was attributed to genes at the KTCN7 locus [55], and SOX21 has been proposed as a candidate for growth hormone deficiency in 13q deletions, although it is absent from many type 2 deletions, including that of a patient with this deficiency [14]. The extent of the deletion must therefore be read before attributing the whole phenotype to MIR17HG [11].

2.6. Distinct mechanisms for overlapping phenotypes

The similarity between the two types led to the proposal that MYCN acts through miR-17~92, but the absence of gastrointestinal atresia in patients with deletion of the cluster and in mice indicates that MYCN controls intestinal development through independent pathways [13]. Mirzamohammadi et al. showed in mice that Mir17-92 deficiency increases TGF-β (transforming growth factor β) signalling, whereas Mycn deficiency reduces PI3K signalling in limb mesenchymal cells; genetic or pharmacological inhibition of TGF-β corrects the skeletal defects of Mir17-92 deficiency but not those of Mycn deficiency, which are partially corrected by Pten heterozygosity [56]. According to Muriello et al., overexpression of miR-17~92 corrects the skeletal anomalies of Mycn-deficient mice in the same work [14]. These results are preclinical and none has been tested in patients [56].

2.7. Gain of dosage: MPAPA and mirror-image duplications

Threonine 58 of MYCN, in the N-terminal domain, is phosphorylated by GSK3B (glycogen synthase kinase 3β), a signal that directs the protein to ubiquitination by the FBXW7 ubiquitin ligase and degradation [12, 40]. The p.(Thr58Met) and p.(Pro60Leu) variants reduce this phosphorylation, increase protein stability even in the presence of FBXW7 and maintain activation of CCND1 and CCND2 in HEK293T cells [12]. Overexpression of Mycn mimicking the variant promotes proliferation and alters neuronal migration in murine corticogenesis [40]. In the T58M/WT model, mice are born in Mendelian proportions and have megalencephaly due to excess proliferation of intermediate precursors, with delayed neuronal migration; more than 90% have postaxial polydactyly, and imperforate vagina with hydrometra and renal tubular anomalies are observed, without tumours [12]. In the same work, mice with a frameshift variant in exon 1 (FS/WT) had lighter brains and a thinner primary motor cortex, due to insufficient proliferation of neural progenitors, in contrast to the T58M/WT mice, which had a thicker cortex; the latter also had larger kidneys, a single kidney or dilatation of the convoluted tubules, and the females were infertile, with uterovaginal malformations [36]. The p.(Thr58Met) variant is frequent as a somatic variant in neuroblastoma, astrocytoma and mesonephric carcinoma, and p.(Pro60Leu), absent from gnomAD, is listed in COSMIC as somatic in basal cell carcinoma [12].

The germline 2p24 duplication including MYCN and DDX1, genes separated by about 340 kb, is associated with neuroblastoma and Wilms tumour; ALK, also implicated in neuroblastoma predisposition, lies about 14 Mb further away, at 2p23.1, and is not always included [23]. ClinGen assigns region ISCA-46287 (chr2:15,568,553–16,045,215, GRCh38) a triplosensitivity score of 2, with limited evidence, for neuroblastoma [7]. For MYCN alone, the triplosensitivity score is 0 [7]. At 13q31.3, Siavrienė et al. described a de novo microduplication including MIR17HG without GPC5 in a woman with developmental delay, skeletal and digital anomalies, tall stature and macrocephaly, and pooled two earlier families with overlapping duplications and a similar phenotype [41].

2.8. Penetrance, expressivity, correlations and modifiers

Penetrance of the major features, especially the digital ones, is apparently complete, but expressivity is highly variable [3]. No significant phenotypic differences have been shown between deletions and missense, nonsense or frameshift variants [3]. In the series by Cognet et al., patients with and without a MYCN variant did not differ in the core features, except for 4–5 toe syndactyly, which was more frequent in the group with a variant [28]. Intrafamilial variability includes monozygotic twin girls with the p.(Arg382His) variant, one with sensorineural hearing loss and the other with duodenal atresia, whose carrier mother and maternal grandfather had only digital anomalies and whose maternal uncle had intellectual disability and structural renal anomalies [27, 57], and families in which one carrier has only microcephaly while another has oesophageal atresia [29]. In the Polish family of Klaniewska et al., the authors did not confirm an association between atresia and a more severe overall phenotype, and the hypothesis of genetic modifiers lacks sufficient evidence [29].

Some severe or atypical phenotypes result from a second diagnosis. Tedesco et al. described the first patient with type 1 and severe intellectual disability who had, in addition to the maternal MYCN variant, a paternal GNAO1 variant [16]. Other examples are hearing loss due to biallelic GJB2 variants [52], Alport syndrome due to COL4A5 in a patient with type 2 [17], variants in SLC26A4 and COMP in a Chinese patient with type 2 [58] and radioulnar synostosis attributed to a SMAD6 variant in a patient with type 1 [59]. The same principle applies to 2p24 deletions, where the extent of the deletion modulates the intellectual disability [15].

2.9. Somatic MYCN alterations: contrast

Somatic MYCN amplification occurs in 20–30% of neuroblastomas and is a marker of poorer prognosis; it is also found in about 25% of alveolar rhabdomyosarcomas, in 5–10% of medulloblastomas (Sonic Hedgehog group, SHH) and in Wilms tumour, retinoblastoma, basal cell carcinoma and lung cancer [12, 19]. The most frequent somatic variant is p.(Pro44Leu), present in 1.7% of high-risk neuroblastomas without amplification [19]. In an older series, Cognet et al. cited amplification in about 10% of neuroblastomas [28], and GeneReviews summarises amplification as present in up to 25% of a diverse set of cancers, always with overexpression [3, 60]. These acquired events do not alter the interpretation of a germline loss-of-function variant, and tumour amplification confers neither actionability nor oncological surveillance on type 1 [9].

3. Clinical Spectrum and Manifestations by System

3.1. Reference cohorts and what each one measures

Published frequencies come from small series, with different inclusion criteria and recruitment biases, and cannot be read as individual risk. The following table summarises the cohorts used throughout the article.

3.2. Type 1: feature frequencies

Sources: [3, 15, 28, 38, 46]. The Celli 2003 values and the Marcelis 2008 values for micrognathia and short stature are cited through the synthesis by Samara et al. [15]. The Oxford Desk Reference gives lower frequencies for oesophageal (25%) and duodenal (20%) atresia, without identifying the source series; the difference reflects the cohort used and not a change in phenotype [20]. Atresia is more frequent in surgically recruited series, such as that of Cognet et al. [28]. The OMIM clinical synopsis gives microcephaly in 79% of cases, syndactyly of toes 2–3 in 56% and of toes 4–5 in 86%, and learning difficulties in 90%, without identifying the denominator [27].

3.3. Type 1: manifestations by system

Hands and feet. Brachymesophalangy (shortening of the middle phalanx of the 2nd and 5th fingers, with clinodactyly of the 5th finger) is the most constant feature, and toe syndactyly affects the 2–3 and/or 4–5 interdigital spaces [3, 20]. Thumb hypoplasia occurs in about one in six patients [3]. Digital expression can be minimal: in a three-generation Polish family with the c.266dup variant, all carriers had microcephaly and toe syndactyly, only one had clinodactyly and none had brachymesophalangy or hypoplastic thumbs [29]. Peleg et al. described congenital absence of the flexor pollicis longus tendon in five affected individuals over three generations of an Israeli family with p.(Arg391Cys) [51]. Ferroul et al. proposed looking for amesophalangy of the toes in the feet [59].

Skull, face and central nervous system. Microcephaly is the second most frequent feature; GeneReviews uses a head circumference below the 10th centile as the threshold [3]. Head circumference may be normal at birth: in the series by Cognet et al. it was normal in three patients at birth and postnatal microcephaly was constant after 3 years of age [28], and the proband of Mérida De la Torre et al. had a head circumference on the 35th centile at birth [46]. In other cases, microcephaly is detectable in the second trimester of gestation [15, 18]. Facial features are subtle: short palpebral fissures, broad nasal root, micrognathia and round face [28]. Cleft palate [46], laryngeal cleft, agenesis of the corpus callosum [59], choanal atresia, frontal balding and focal epilepsy [61], and white matter hyperintensities on T2-weighted imaging [46] have been described.

Gastrointestinal tract. Atresia is the main medical concern and requires immediate surgery [3]. Oesophageal atresia, with or without tracheoesophageal fistula (TOF), and duodenal atresia are the dominant forms, isolated or combined [3, 32]. Holder-Espinasse et al. described duodenal atresia in a father and his two children [62]. Distal intestinal atresia was first described by Büttiker et al. in an Australian family: the daughter had long-gap oesophageal atresia without fistula (type A) and the father, with short stature and learning difficulties, had imperforate anus with rectovesical fistula [63]. The syndrome is described as the most frequent cause of familial syndromic gastrointestinal atresia [38], and in a series of 148 cases of oesophageal atresia from Necker-Enfants Malades it was among the associated syndromes [64].

Neurodevelopment and behaviour. Mild learning disability predominates; overt intellectual disability is rare, but mean intelligence is lower than that of the population and of unaffected relatives, and most adults live independently [3]. There are exceptions: a boy from Kosovo with severe intellectual disability and a novel variant [65], the patient of Tedesco et al. with an additional GNAO1 variant [16] and the 9-year-old brother in the family of Klaniewska et al., without atresia, who walked at 3 years and spoke at 5, with hyperactivity and attention deficit [29].

Growth. Short stature below the 3rd centile is uncommon, but mean height is lower than that of the population [3]. In the review by Shaw-Smith et al., of the 18 cases with recorded height, 3 (16.7%) were at or below the 0.4th centile and 9 (50%) at or below the 10th centile; OMIM summarises the same review as 4 cases at or below the 3rd centile and 9 at or below the 10th centile, with the father and son described both having short stature [27, 66]. The father in the Málaga family had short stature treated with growth hormone [46].

Kidneys and urinary tract. Renal anomalies, present in less than half of patients, include horseshoe kidney, dysplastic kidneys, hydronephrosis and pelvic dilatation, chronic nephritis and vesicoureteral reflux with dysplasia and renal failure [3]. Aslam et al. described a girl with reflux nephropathy and end-stage kidney disease at 6 years of age [67], and Celli et al. recorded a patient with hydronephrosis of one kidney and cystic dysplasia of the other, who underwent nephrectomy [35].

Heart. The heart defects described include patent ductus arteriosus, multiple ventricular septal defects (VSD), tricuspid stenosis, VSD with tricuspid atresia and interrupted aortic arch [3]; Cognet et al. observed atrial septal defect and VSD [28].

Hearing. Hearing loss, more often sensorineural than conductive, affects 7–10% of patients [3, 59]. Inner ear malformation is relevant for rehabilitation: Dodds et al. described severe stenosis of the internal auditory canals with absent auditory nerves, which ruled out cochlear implantation [68], and Chen et al. found cochlear nerve hypoplasia with internal auditory canal stenosis and dilated vestibules in a 4.4 Mb deletion [21].

Other. Vertebral anomalies have been described [35, 69], including, in a family of a father and two children, a sagittal cleft of the S4 body with absent S5 in one child and a C5–C7 block vertebra in the father [27], as well as asplenia and diaphragmatic hernia [28]. The OMIM clinical synopsis adds a small anterior fontanelle, prominent occiput, triangular face, facial asymmetry, low-set posteriorly angulated ears, epicanthus, upslanting palpebral fissures, high palate, thumb symphalangism, vocal cord paralysis, annular pancreas, accessory spleen, congenital asplenia and, in the prenatal period, decreased fetal movements and polyhydramnios [27]. A patient with the p.(Glu73Ter) variant had chronic nephritis, renal failure and hypertension [27]. A 12-year-old girl with features of MMT and of Rett syndrome was proposed to have a contiguous gene syndrome, a hypothesis that OMIM considers unlikely because the two loci are on different chromosomes [27].

3.4. Contiguous gene deletions at 2p24

Of the six individuals of Burnside et al., five had clinical data: core type 1 features (microcephaly, digital anomalies, gastrointestinal atresias) with structural heart disease, hearing loss and renal anomalies, as well as motor delay, structural brain anomalies, genital anomalies and radioulnar synostosis, which the authors did not specifically associate with type 1 [5]. The 26-month-old girl of Chen et al. had profound bilateral hearing loss, microcephaly, psychomotor delay, short palpebral fissures, hypertelorism, brachymesophalangy of the 2nd and 5th fingers and bilateral clinodactyly [21]. Small deletions, from 342 kb to 1.5 Mb, may give mild pictures identified only after a prenatal diagnosis in the fetus [15, 18].

3.5. Type 2

The table compiled by Demir et al. in 2022 is the most complete synthesis of type 2 [17].

In the literature, brachymesophalangy was present in 16/16 (100%), short stature in 13/16 (81%) and clinodactyly of the 5th finger in 11/16 (68%) [17]. Published counts diverge according to the inclusion criterion: 10 patients up to 2015 [70], 12 published cases up to April 2017 [14], six according to Sirchia et al. in 2017 [55], fewer than 25 up to 2021 [58] and 23 including DECIPHER in 2022 [17]. A 2026 compilation attributes hearing loss in 66% to type 2, a figure not supported by the source series, whose three patients had normal or unreported audiology; the table of Demir et al. (2 of 23, with 15 without information) is the one supported by the primary sources [14, 17, 46].

Type 2 adds its own features. Muriello et al. described in three patients growth hormone deficiency with compression of the adenohypophysis, aortic root dilatation, bicuspid aortic valve, interphalangeal joint contractures, chronic musculoskeletal pain, Raynaud phenomenon, distal sclerodactyly, memory problems and sleep disturbances [14]. Heart disease was present in 50% of patients with a reported cardiac examination, including bicuspid aortic valve with stenosis and tetralogy of Fallot in earlier reports [14, 70]. Grote et al. described the first paediatric patient without microcephaly, a 14-year-old boy with bicuspid aortic valve and aortic stenosis, hearing loss and mild learning difficulties [70]. Ganjavi et al. described a psychiatric presentation with severe anxiety, obsessions, skin-picking disorder (dermatillomania) and self-harm [71]. Developmental dysplasia of the hip and thumb polysyndactyly [17, 58], bilateral keratoconus in a 58-year-old woman with a deletion including the KTCN7 locus [55] and markedly hypoplastic thumbs in two patients from the founding series [13] have also been described. Metacarpophalangeal pattern profile analysis was applied to a 3-month-old infant with type 2 [72]. In the founding series, the two probands had mild intellectual disability, the fathers were equally affected and one of them had an affected sister [53]. OMIM classifies type 2 as a contiguous gene deletion syndrome involving MIR17HG [53].

3.6. MPAPA

Source: [12]. The OMIM clinical synopsis describes in the first patient postaxial polydactyly of the hands and feet, epicanthus, hypertelorism, high palate, long philtrum, prominent and high forehead, posteriorly rotated ears and thick, laterally extended eyebrows, and records eventration of the diaphragm, tracheomalacia, Meckel diverticulum and pelviureteric dilatation, each in a single patient [6]. In patient 1, second-trimester ultrasound showed ventriculomegaly and postaxial polydactyly, fetal magnetic resonance imaging (MRI) at 30 weeks showed macrocephaly with biparietal and fronto-occipital diameters above the 95th centile, and autopsy revealed a Meckel diverticulum, left pelviureteric dilatation and brain weight above the 95th centile [12]. In patient 2, a tumour detected at birth was classified at 3 months as a Children's Oncology Group intermediate-risk neuroblastoma, with lesions in both adrenal glands, the liver and a prevertebral location [12]. The authors describe a recognisable face, with hypertelorism, broad and depressed nasal root, broad nasal base and upturned nasal tip, and highlight the hypoplastic corpus callosum despite the megalencephaly as a distinctive radiological feature [12].

3.7. Mirror-image duplications

The 13q31.3 microduplication including MIR17HG associates tall stature and macrocephaly with developmental delay and skeletal and digital anomalies, the inverse of type 2 [41]. The 2p24 duplication including MYCN and DDX1 manifests mainly as predisposition to neuroblastoma and Wilms tumour [23].

3.8. Natural history and treated disease

The sources include no mortality or survival series; information on outcomes comes from reports of cases that had already been treated. In the family of Klaniewska et al., the two siblings with type C oesophageal atresia repaired thoracoscopically on day 2 of life developed anastomotic stricture treated with 2 and 3 endoscopic dilatations and, at 5 and 3 years of age, ate without difficulty or with only mild difficulty with solids [29]. The adults with type 2 described reached the seventh and eighth decades, one of them working as an accountant at 70 years of age [14, 55]. End-stage kidney disease in childhood is a possible complication of unrecognised vesicoureteral reflux [67]. In MPAPA, the two neuroblastomas treated with chemotherapy and surgery were in remission [12].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age

Prenatal period. Polyhydramnios with a small or absent stomach suggests oesophageal or duodenal atresia, and the Oxford Desk Reference recommends considering a monogenic cause, such as Feingold syndrome, when there is a family history of intestinal atresia [20, 73]. Isolated fetal microcephaly in the second trimester, with or without bilateral clinodactyly, led to the diagnosis of 2p24 deletions in two recent reports, both inherited from parents with undiagnosed mild features [15, 18]. The combination of ventriculomegaly, macrocephaly and postaxial polydactyly is the prenatal pattern of MPAPA [12].

Newborn. A newborn with oesophageal atresia, tracheoesophageal fistula or duodenal atresia should have the hands and feet examined for short 5th fingers with clinodactyly, brachymesophalangy of the 2nd and 5th fingers and toe syndactyly, and the head circumference measured [3, 20]. Oesophageal atresia is associated with other anomalies in about half of cases, and Feingold syndrome is one of the causes to consider in a tracheoesophageal fistula associated with other anomalies [74, 75].

Child. The association of microcephaly, learning difficulties and digital anomalies, without atresia, suggests either type 1 or type 2 [3, 4]. Marked short stature, especially with growth hormone deficiency, bicuspid aortic valve or aortic root dilatation, points to type 2 [14, 70]. Sensorineural hearing loss in a child with microcephaly and short 5th fingers is another mode of presentation [21, 52]. A child with megalencephaly, postaxial polydactyly and neuroblastoma in infancy should raise the possibility of MPAPA [12].

Adult and parents. Because about 60% of those diagnosed have an affected parent and expression can be minimal, the parents of a proband should be examined, including hand radiographs, before the family history is considered negative [3, 9, 16].

4.2. Warning and emergency signs

Oesophageal atresia requires immediate care: inability to pass a nasogastric tube of 10 Fr or larger confirms the suspicion, and placement of a Replogle tube in the upper oesophageal pouch with continuous low-pressure suction prevents aspiration before surgery [76]. When suspected prenatally, delivery should take place in a hospital with neonatal intensive care and paediatric surgery, with a senior paediatrician present [20]. After repair, food aversion, food bolus impaction, odynophagia, choking, cough, pneumonia, vomiting and malnutrition suggest dysphagia and require investigation [77]. Stridor or swallowing disorders raise the possibility of laryngeal cleft [59], and life-threatening events require exclusion of anatomical causes and aspiration before any surgery [77]. Recurrent urinary tract infections or hydronephrosis require investigation for vesicoureteral reflux, given the possibility of end-stage kidney disease [67]. In MPAPA and 2p24 duplication, the family should know the signs of neuroblastoma: abdominal mass or pain, opsoclonus-myoclonus or ataxia, subcutaneous nodules, watery diarrhoea, abdominal distension and periorbital ecchymosis [23].

4.3. Newborn screening

There is no newborn screening for any of the entities in scope; recognition depends on examination of the newborn with atresia and on prenatal ultrasound [3, 20].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

There are no consensus-validated diagnostic criteria; the published criteria are series inclusion criteria or author proposals [3].

The diagnosis of type 1 is established in a proband with suggestive findings and a heterozygous pathogenic or likely pathogenic MYCN variant; a variant of uncertain significance (VUS) neither establishes nor excludes the diagnosis [3, 78]. Type 2 is established by a 13q31.3 deletion including MIR17HG, and MPAPA by a MYCN variant with demonstrated gain of function [11, 12].

5.2. Non-molecular clinical and imaging assessment

There is no biochemical biomarker and no non-molecular confirmatory test [9]. Assessment includes radiographs of the hands and feet (brachymesophalangy of the 2nd and 5th fingers, thumb hypoplasia, amesophalangy of the toes), head circumference plotted on reference charts, dysmorphological examination, prenatal and postnatal imaging of the gastrointestinal tract, renal ultrasound, echocardiography, and audiological and developmental assessment [9, 59]. Hand radiographs of the parents help to assess inheritance [9]. In the approach to an atresia, the Oxford Desk Reference recommends recording the family history, maternal age, maternal diabetes and teratogen exposure, examining the spine, upper limbs, digits, ears and eyes, requesting a spinal radiograph in oesophageal atresia, chromosome breakage studies for Fanconi anaemia when there are bilateral upper limb anomalies, echocardiography, renal ultrasound and ophthalmological assessment, storing DNA and, in jejunoileal or colonic atresias, testing for cystic fibrosis [20].

5.3. Molecular tests and algorithm

GeneReviews describes two routes. When the phenotype is typical, MYCN sequencing is performed (which detects small insertions and deletions and missense, nonsense and splice variants, but not exon or whole-gene deletions or duplications) and, if negative, targeted deletion/duplication analysis by quantitative PCR, long-range PCR, multiplex ligation-dependent probe amplification (MLPA) or targeted microarray; alternatively, a multigene panel including MYCN and copy-number analysis is used, bearing in mind that some intellectual disability panels do not include MYCN [3]. When the phenotype is atypical, exome or genome sequencing is performed and, if the exome is not diagnostic, an exome array for deletions and duplications of one or more exons [3]. The 2021 guideline of the American College of Medical Genetics and Genomics (ACMG) strongly recommends exome or genome sequencing as a first- or second-tier test in patients with one or more congenital anomalies before 1 year of age or with developmental delay or intellectual disability before 18 years, after CMA or targeted testing according to clinical judgement, and keeps targeted testing first when the presentation strongly suggests a specific diagnosis; the recommendation is based on a systematic review of 167 studies and an Ontario health technology assessment [22, 79]. In that assessment, the diagnostic yield was 38% with extended genomic sequencing versus 21% with standard testing (risk ratio 1.76; 95% confidence interval, 95% CI, 1.20–2.58), 43% with genome and 34% with exome [22].

5.4. Yield and limitations of tests

The GeneReviews values are proportions of probands with a variant detected by each method and not clinical sensitivities [3].

The total of 68% suggests that about one third of patients with a clinical diagnosis in the series underlying the table had no MYCN variant [3]. The primary data are consistent with this reading: Cognet et al. found a variant or deletion in 8 of 17 patients (47%) and none in 12 patients with isolated oesophageal atresia [28], de Pontual et al. estimated that MYCN explained about 70% of families [13], and two of the ten index cases without a MYCN variant had the type 2 deletion [13]. Unexplained cases may correspond to type 2, phenocopies or mechanisms not yet described; sequencing of five conserved non-coding elements and a high-density microarray of 3.3 Mb around MYCN were negative [28]. Exome sequencing poorly detects intronic variants outside exon-intron junctions, repeat expansions, methylation changes and many CNVs [22]. MIR17HG deletions are not detected by panel sequencing or by exome sequencing without CNV calling, because the gene is non-coding [11]. A deletion identified by CMA must be delineated to determine whether it includes SOX11 or, at 13q, GPC5, GPC6 and other genes [11, 15]. No case of germline mosaicism has been described, but the parent in whom the variant arose may have somatic mosaicism and be mildly affected [3]. A negative result does not exclude the clinical diagnosis.

5.5. Report, variant interpretation and segregation

The report should state the gene, the transcript and its version (NM_005378.6 for MYCN), the HGVS c. and p. notation, the zygosity, the class according to ACMG/Association for Molecular Pathology (AMP) and, for CNVs, the genomic coordinates with the genome build and the genes included [10, 78]. The pathogenic and likely pathogenic classes have the same clinical value [3, 78]. Missense variants outside the bHLH domain require additional data: the Málaga family shows how co-segregation, rarity and phenotype specificity can coexist with benign in silico predictions and leave the variant as a VUS [46]. Because expressivity is highly variable, segregation should combine genotype with clinical and radiographic examination, since several mildly affected relatives were identified only after testing [16]. The ClinGen haploinsufficiency score of 3 supports classification of MYCN deletions as pathogenic [7, 15]. DNA banking is recommended when the molecular diagnosis is not established [3, 80].

5.6. Prenatal and preimplantation diagnosis

In the 18 European congenital anomaly registries analysed by Haeusler et al., about one third of gastrointestinal obstructions were detected before birth, 40% of these before 24 weeks [20, 81]. Second- and third-trimester ultrasound shows polyhydramnios and a small or absent stomach [20]. In fetal microcephaly, defined as a head circumference 2–3 standard deviations below the mean or a head circumference/abdominal circumference ratio below the 3rd centile, Gouveia et al. propose clinical and family history, detailed ultrasound, TORCH serology (toxoplasmosis, syphilis, parvovirus, varicella-zoster, Zika, rubella, cytomegalovirus, herpes and human immunodeficiency virus) and drug review, invasive infectious and genetic testing and fetal MRI, with amniocentesis ideally between 15+0 and 17+6 weeks and possible after 24 weeks with a slightly higher risk [18]. In the two prenatal reports, the diagnosis came from CMA on amniotic fluid, after negative quantitative fluorescent PCR (QF-PCR) and cytomegalovirus testing in the Portuguese case [15, 18]. When the familial variant is known, prenatal diagnosis and preimplantation genetic testing are possible, and any pregnancy with a carrier fetus or a 50% risk should have high-resolution ultrasound with fetal echocardiography [3, 18].

6. Differential Diagnosis

6.1. Comparative table

AR, autosomal recessive.

6.2. Main differentials

The differential with the VACTERL association is the most frequent in practice, because both combine oesophageal or anal atresia, heart disease and renal anomalies. GeneReviews distinguishes type 1 by microcephaly, brachymesophalangy and toe syndactyly, and notes that the molecular basis of VACTERL is unknown [3]. VACTERL is a diagnosis of exclusion and has no established diagnostic criteria [20, 35]. In a series of 68 patients with VACTERL from Erasmus MC, no de novo CNVs larger than 30 kb were found, and the authors included Feingold syndrome, Townes-Brocks syndrome and Fanconi anaemia in the differential [82]. The Brazilian textbook of genetic syndromes lists the same differences (brachymesophalangy, toe syndactyly, microcephaly, cognitive impairment and facies) and estimates that about 90% of VACTERL cases are sporadic, with a negligible recurrence risk, in contrast with the 50% of type 1 [83]. The distinction therefore has direct consequences for counselling.

Fanconi anaemia shares thumb hypoplasia, microcephaly, growth restriction, intestinal or anal atresia and renal anomalies, but has no brachymesophalangy or toe syndactyly and is associated with a tumour risk that does not exist in type 1 [3]. In the presence of bilateral upper limb anomalies, chromosome breakage studies are indicated [20]. CHARGE syndrome is distinguished by coloboma and by ear and genital anomalies [3]. In the series of 148 cases of oesophageal atresia from Necker-Enfants Malades, 9 (6%) had a known genomic cause, 22 (14%) VACTERL or VACTERL with oculoauriculovertebral spectrum and 65 (44%) other malformations without a known syndrome, and the syndromes cited as most frequent were CHARGE, EFTUD2 mandibulofacial dysostosis, Feingold syndrome, trisomy 18 and Fanconi anaemia [64].

Brachydactyly type A4, or brachymesophalangy of the 2nd and 5th fingers (ORPHA:93394; OMIM 112800), has a digital pattern very similar to that of type 1 [3]. No MYCN variants have been found in families with BDA4, but GeneReviews allows for testing MYCN in these families [3]. Zhao et al. associated a HOXD13 polyalanine contraction with a phenotype overlapping brachydactylies A4, D and E with mild 2–3 toe syndactyly [84]. Isolated brachymesophalangy of the 5th finger (brachydactyly type A3) is a common normal variant, present in 24.5% of Japanese women, and should not be overinterpreted [89]. Other causes of brachymesophalangy include autosomal dominant Robinow syndrome type 2 due to DVL1, Werner-type mesomelic dysplasia with hypoplastic thumbs, ischiopubic-patellar dysplasia, Apert, Pfeiffer, Carpenter and Poland syndromes and multiple synostoses syndrome [86, 90–92]. The differential of the digital sign also includes orofaciodigital syndrome type 18 (ORPHA:508501; OMIM 617927), with short stature and brachymesophalangy [93].

The differential of duodenal atresia begins with trisomy 21, responsible for 30–50% of cases, and includes maternal diabetes with caudal regression, the OEIS complex (omphalocele, exstrophy, imperforate anus and spinal defects), vascular disruption, Martínez-Frías, Mitchell-Riley and Fryns syndromes, alveolar capillary dysplasia due to FOXF1 and laterality disorders [20]. In jejunoileal atresia, cystic fibrosis must be excluded: in one series, 4 of 38 Caucasian infants with jejunoileal atresia had cystic fibrosis [20]. Carbimazole and methimazole (thiamazole) in pregnancy are teratogens associated with oesophageal atresia [20]. The Brazilian textbook adds Alagille, Baller-Gerold, Currarino, 22q11.2 deletion, Fryns, Holt-Oram, MURCS, Goldenhar, Opitz G/BBB, Pallister-Hall and Townes-Brocks syndromes to the differential of VACTERL [83].

The distinction between types 1 and 2 is clinical (gastrointestinal atresia only in type 1) and molecular, and de Pontual et al. therefore recommended testing MYCN and MIR17HG in all patients with the skeletal features of the syndrome [13]. In MPAPA, the main differential is MPPH3 due to gain of function of CCND2, a MYCN target: perisylvian polymicrogyria, hydrocephalus and a thick corpus callosum point to CCND2, and a hypoplastic corpus callosum with neuroblastoma points to MYCN [12]. Disorders of the SHH and Wnt pathways with polydactyly, such as Greig and Pallister-Hall syndromes due to GLI3, postaxial polydactyly A1/B, Robinow syndrome and Fuhrmann syndrome, also enter the differential of postaxial polydactyly [19]. A patient with exudative retinopathy, cerebral calcifications, duodenal atresia, preaxial polydactyly, micropenis, microcephaly and short stature, with normal MYCN and miR-17~92 and no alterations in CTC1 or the telomeres, was described as a possible new syndrome [88]. The Feingold eponym also applies to Floating-Harbor syndrome, unrelated to this class [30].

7. Treatments and Risk Management Strategies

7.1. Principles and organisation of care

None of the entities in scope has a disease-modifying treatment, a dedicated guideline or a registered clinical trial; management is organ-directed, by a multidisciplinary team, and is based on GeneReviews and on the guidelines for the associated malformations [3, 9]. Atresias require surgical treatment, anomalies of the fingers and toes are treated with occupational therapy or surgery, cardiac and renal anomalies follow usual practice, significant hearing loss is rehabilitated, and development is monitored with educational intervention [3]. For oesophageal atresia, the 2020 consensus of the European Reference Network for rare Inherited and Congenital Anomalies (ERNICA) unanimously recommends referring the pregnancy with suspected atresia for multidisciplinary prenatal counselling at a specialist centre, operating on and following patients at centres with a multidisciplinary team and a minimum mean caseload of 5 new cases per year, with follow-up until transition, and involving parents in patient associations as early as possible [94]. In a survey of ERNICA and affiliated centres conducted between December 2020 and January 2021, 29 of 32 centres from 17 countries responded: 79% treated more than 5 new cases per year and two thirds had a dedicated multidisciplinary clinic, accessible to all patients in 60% of centres [95].

7.2. Oesophageal atresia: neonatal period and surgery

When oesophageal atresia is suspected prenatally, delivery should be planned in a hospital with full neonatal intensive care and links to specialist paediatric surgery, with a senior paediatrician present to reduce the risk of aspiration pneumonia [20]. In isolated tracheoesophageal fistula, surgical division of the fistula is curative [20]. The recommendations of the 2020 ERNICA consensus for oesophageal atresia with fistula and primary anastomosis, voted on a scale of 1 to 9 with consensus defined as a score of 6–9 by at least 75% of voters, are as follows [76].

Only 4 of the 52 recommendations (7.7%) had level 1 evidence according to the Oxford Centre for Evidence-Based Medicine, and most are based on expert opinion [76]. Renal ultrasound before discharge is of particular interest in type 1, given the risk of reflux and renal dysplasia [3, 67].

7.3. Long-gap oesophageal atresia

Long-gap atresia, such as that of the daughter in the family of Büttiker et al. [63], has its own 2021 ERNICA consensus, with 97 items voted on, full consensus on 56 (58%) and consensus of at least 75% on 90 (93%) [96]. The definition adopted is any atresia with no air in the abdomen or with a gap of at least three vertebral bodies [96]. When oesophageal replacement is needed, the preferences were gastric transposition (39%), jejunal interposition (17%) and colonic interposition (13%); cervical oesophagostomy (with full consensus) and the Kimura technique are not recommended, nor is postoperative antibiotic prophylaxis for more than 48 hours [96]. After reconstruction, the consensus recommends clinical and nutritional assessment every 3 months in the first year and then at least every 2 years until transition, upper gastrointestinal endoscopy and/or pH-impedance monitoring one year after reconstruction after tapering the proton pump inhibitor (PPI), at least two further endoscopies until transition, respiratory review by respiratory medicine, bronchoscopy only if there are symptoms, screening for dumping in patients with microgastria, pyloroplasty or antireflux surgery, a specific transition programme and quality-of-life assessment with a validated instrument [96].

7.4. Duodenal atresia and distal atresias

Duodenal, jejunal or anal atresia is treated surgically according to usual paediatric surgical practice, and GeneReviews indicates no technical particularities in type 1 [3]. The sources consulted do not describe series of surgical outcomes of these atresias in the syndrome; the available reports are familial and descriptive [62, 63].

7.5. Gastrointestinal follow-up after oesophageal atresia repair

Acid suppression. The 2016 joint guideline of the European Society for Paediatric Gastroenterology, Hepatology and Nutrition and the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN-NASPGHAN) recommends treating gastro-oesophageal reflux (GOR) with acid suppression in all patients with oesophageal atresia in the neonatal period, with PPIs as first line, and maintaining systematic treatment to prevent peptic complications and anastomotic stricture until the first year of life or longer, depending on persistence of reflux disease; the three recommendations have low-quality evidence and the guideline gives no doses [77]. The 2020 ERNICA consensus agreed with routine acid suppression (88.9%) and with the use of PPIs (100%), recommended gradual tapering at the end of prophylaxis (94.4%), but did not reach consensus on a duration of 12 months (66.7%), because two studies cited by the authors, with 63 and 128 patients, showed no reduction in strictures with prolonged prophylaxis [94]. In the practice of ERNICA centres, 86% prescribed PPIs after repair, with a median duration of one year and a range from less than 6 months to more than 2 years [95]. The omeprazole dose follows the label, by extrapolation: use in newborns under 1 month of age and prolonged prophylaxis after repair are outside the approved indications [77, 97].

The delayed-release oral suspension can be given orally or via a nasogastric or gastric tube, before meals [97].

Reflux monitoring. pH monitoring is useful for assessing severity and the relationship of acid reflux to symptoms, and pH-impedance in selected cases; endoscopy with biopsies is mandatory in monitoring reflux disease; and all patients, even asymptomatic ones, should be monitored when antacid therapy is stopped and during long-term follow-up [77]. ESPGHAN-NASPGHAN recommends three routine endoscopies in asymptomatic patients during childhood: one after stopping the PPI, one before 10 years of age and one at transition to adulthood, because mucosal changes are found in up to 35% of patients without symptoms [77]. The 2020 ERNICA consensus recommended endoscopy at one year (83.3%) and at least two more until transition (100%), scheduled pH monitoring (93.8%) and monitoring when the antacid is stopped (83.3%), with no consensus on two additional pH studies (55.6%), and scheduled pulmonary function tests (77.8%); routine contrast studies and bronchoscopy were not recommended [94].

Anastomotic stricture. A clinically relevant stricture requires narrowing with functional impairment; there is no evidence that routine screening and dilatation are superior to treating symptomatic patients [77]. Diagnosis is by contrast study and/or endoscopy, and dilatation is performed with a balloon or a semi-rigid bougie over a guidewire, under endoscopic or fluoroscopic control and under general anaesthesia with tracheal intubation, repeated at 2–4 weeks if symptoms persist [77, 94]. Recurrent stricture is defined as three or more recurrences requiring dilatation, and the ERNICA consensus considers fundoplication after a maximum of five dilatations (83.3%) [77, 94]. As adjuvants in recurrent strictures, ESPGHAN-NASPGHAN allows intralesional and/or systemic corticosteroids, topical mitomycin C, stents and endoscopic knife therapy; the 2020 ERNICA consensus approved stents and indwelling balloons (100%), but not mitomycin C (26.7%) or corticosteroids (46.2%), whereas the 2021 consensus for long-gap atresia accepted both (77.8% and 94.4%) [77, 94, 96]. The sources consulted give no doses or concentrations for these adjuvants, and mitomycin C has shown inconsistent results depending on concentration [94].

Fundoplication. Potential indications include recurrent anastomotic strictures, reflux disease poorly controlled with maximum-dose PPIs, prolonged dependence on transpyloric feeding and cyanotic spells [94]. Severe dysmotility predisposes to complications after fundoplication, and a barium study, endoscopy with biopsies and pH monitoring are performed beforehand [77].

Eosinophilic oesophagitis, dysphagia and vascular malformations. Eosinophilic oesophagitis should be excluded, with multiple biopsies above and below the anastomosis, at any age when there is dysphagia, reflux, cough, choking or a PPI-refractory stricture, before antireflux surgery, and is treated as in the general population [77]. All patients with dysphagia should have a contrast study and oesophagoscopy with biopsies; manometry is useful, and in the presence of dysphagia, dyspnoea or cyanosis vascular malformations should be excluded by CT angiography or MR angiography [77]. Congenital oesophageal stenosis is investigated by oesophagogram and endoscopy, with endoscopic dilatation as first line [77].

Airway and nutrition. Regular assessment by a multidisciplinary team including respiratory medicine and otorhinolaryngology is performed even in the absence of symptoms; with respiratory symptoms, laryngeal cleft, vocal cord paralysis, recurrent or missed fistula, anastomotic stricture, congenital stenosis and vascular ring are excluded, and acid suppression is used with caution for extraoesophageal manifestations [77]. The laryngeal cleft described in type 1 makes this assessment especially relevant [59]. Feeding disorders require a multidisciplinary approach, and comorbidities increase the risk of malnutrition [77].

7.6. Kidneys, heart, hearing, hands and feet

Renal and cardiac anomalies are treated according to usual paediatric nephrology and cardiology practice [3]. Vesicoureteral reflux deserves attention, because reflux nephropathy can progress to end-stage kidney disease in childhood, and Aslam et al. recommend urinary tract imaging in all children with suspected syndrome [67]. Significant hearing loss is treated according to the standards for hereditary hearing loss [3]. Before considering a cochlear implant, computed tomography (CT) or MRI of the inner ear and cochlear nerve is essential, because agenesis of the auditory nerves can make it impossible [68]; in the patient of Nitta et al., with cochlear nerve canal stenosis and vestibular enlargement, the implant improved hearing thresholds, although language and social delays persisted [52]. Digital anomalies are assessed with a hand surgeon, an occupational therapist and a foot specialist, according to functional needs [3]. Ferroul et al. recommend checking flexion of the thumb interphalangeal joint to look for absence of the flexor pollicis longus tendon [51, 59].

7.7. Neurodevelopment, education and behaviour

GeneReviews describes a North American model, which it acknowledges varies between countries: early intervention from 0 to 3 years (occupational therapy, physiotherapy, speech and feeding therapy, infant mental health, special education and sensory specialists), preschool with an individualised education plan between 3 and 5 years, follow-up by a developmental paediatrician at all ages, a plan reviewed annually in the least restrictive environment possible, vision and hearing consultants on the team and transition planning in adolescence [3]. The message about cognitive prognosis requires care, because it is easy to overestimate or underestimate it [9]. In type 2, Muriello et al. recommend attention to mood, memory and sleep [14]; one of their patients had attention deficit hyperactivity disorder (ADHD) treated with unspecified pharmacotherapy, and mouse models show memory deficits and neurotransmitter alterations [14, 54].

7.8. Type 2: growth hormone deficiency and aortic surveillance

In patient 1 of Muriello et al., brain MRI requested at 2 years for short stature showed the adenohypophysis compressed by a probable Rathke's cleft cyst, with a partially empty sella and normal neurohypophysis; at 3 years, the peak growth hormone (GH) on stimulation testing was 2.2 ng/ml, and somatropin was started at a weight-adjusted dose [14]. Growth velocity, below the 3rd centile before treatment, rose above the 75th centile between 3.5 and 5.5 years and fell to or below the 50th centile after 6 years, with high-normal insulin-like growth factor 1 (IGF-1) and a plateau at 8 years with no secondary cause; at 11 years he was on the 1st centile for height, against a mid-parental target on the 75th centile [14]. Two patients with type 2 previously treated with GH had no clear benefit in height, and in neither of them was the pituitary imaged [14, 55, 70]. The authors recommend referral to endocrinology and testing for GH deficiency in any patient with type 2 and short stature [14]. The 2016 guideline of the Pediatric Endocrine Society (PES) on GH and IGF-1 guides this assessment [98]. It waives stimulation testing only when auxological criteria, a hypothalamic-pituitary defect (major congenital malformation, tumour or irradiation) and deficiency of at least one other pituitary hormone coexist; in isolated deficiency, as in the patient of Muriello et al., it recommends not using the tests as the sole criterion (strong) and requires an inadequate response to two different tests, which can be performed sequentially on the same day [14, 98]. The usual threshold of 10 µg/L has no controlled reference standard and had an estimated specificity of 25% in the North American NCGS registry, whereas very low peaks (below 5 µg/L) are associated with severe deficiency and a better response; the peak of 2.2 ng/ml (2.2 µg/L) in the patient of Muriello et al. lies within this range, close to the median of 2.7 µg/L observed in 1,071 patients in the GeNeSIS registry with imaging abnormalities [14, 98]. The PES recommends that laboratories use assays harmonised with the somatropin standard IRP IS 98/574 (strong), because differences between assays misclassified up to 29% of patients; it notes that the response decreases with excess weight and that hypothyroidism and hypercortisolism should be excluded before testing; and it suggests sex steroid priming in prepubertal boys over 11 years and girls over 10 years with a predicted adult height within −2 SD (2 mg oral β-oestradiol, 1 mg below 20 kg, on the two preceding nights, or, in boys, 50–100 mg depot testosterone IM one week before) [98].

Somatropin, recombinant human growth hormone, is approved for paediatric GH deficiency, which is the indication to use when the deficiency is demonstrated; Feingold syndrome is not, in itself, an indication [99, 100]. The US Genotropin® label recommends 0.16–0.24 mg/kg/week subcutaneously, with the weekly dose divided into 6 or 7 injections and rotation of sites to avoid lipoatrophy, and discontinuation when the epiphyses close; lack of response in the first year requires review of adherence and exclusion of hypothyroidism, undernutrition, advanced bone age and anti-GH antibodies [99]. The European summary of product characteristics of Omnitrope® recommends 0.025–0.035 mg/kg/day or 0.7–1.0 mg/m²/day, with continuation in adolescence until somatic development is complete when the deficiency persists, and notes that higher doses have been used [100]. Neither label defines a maximum dose in paediatric GH deficiency [99, 100]. The two dosing regimens are equivalent in practice (0.16–0.24 mg/kg/week corresponds to about 0.023–0.034 mg/kg/day) [99, 100]. Contraindications include acute critical illness, active neoplasia (any previous neoplasia must be inactive with treatment completed), proliferative diabetic retinopathy, hypersensitivity and closed epiphyses; before starting, a pituitary or brain tumour should be excluded, because GH deficiency may be its first sign, and during treatment the progression of pre-existing tumours, the risk of second neoplasia in cancer survivors, glucose intolerance, intracranial hypertension (fundoscopy looking for papilloedema) and fluid retention are monitored [99]. Pituitary MRI before treatment has, in type 2, the dual justification of the label and of the cyst described by Muriello et al. [14, 99]. The PES guideline dose is the reference and the labels are shown for comparison: starting dose of 0.16–0.24 mg/kg/week (22–35 µg/kg/day), by weight or body surface area, 6–7 days per week, individualised thereafter (strong), the same as the Genotropin® label and with a lower limit below the 0.025 mg/kg/day of the Omnitrope® SmPC [98–100]. The PES suggests measuring IGF-1 to monitor adherence and lowering the dose if it exceeds the laboratory range for age or pubertal stage (conditional), recommends not routinely increasing to 0.7 mg/kg/week at puberty (strong) and not continuing the paediatric dose once growth velocity falls below 2–2.5 cm/year (strong), a more restrictive criterion than continuation until complete somatic development in the Omnitrope® SmPC [98, 100]. It recommends informing about and monitoring, at each visit, intracranial hypertension (about 28 per 100,000 treatment-years), slipped capital femoral epiphysis (73 per 100,000 treatment-years) and scoliosis progression (about 0.2%), reassessing the adrenal and thyroid axes after starting GH when the cause may involve other pituitary deficiencies, and suggests not routinely performing cardiac assessment, densitometry or lipid profile; in type 2, echocardiography is justified by the bicuspid aortic valve and aortic dilatation, not by GH [14, 98]. In conditions with an intrinsically increased risk of neoplasia, the PES recommends counselling about the lack of evidence regarding the effect of GH on that risk; the guideline does not mention MPAPA or 2p24 duplication, and applying this recommendation to these entities, with predisposition to neuroblastoma, is an extrapolation [23, 98].

Aortic root dilatation in the same patient was detected at 32 months (root 1.88 cm, Z +3.49; annulus 1.32 cm, Z +3.02) and persisted at 10 years (root 2.65 cm, Z +3.11; annulus 1.83 cm, Z +2.60; sinotubular junction 2.11 cm, Z +2.86) [14]. Because bicuspid aortic valve is associated with aortic dilatation or aneurysm in 20–84% of cases in the general population, the authors recommend echocardiography with assessment of the aortic valve and ascending aorta at diagnosis in any patient with type 2 [14]. The sources consulted give no pharmacological treatment for the aortic dilatation and no reassessment intervals specific to this syndrome; the patient was reassessed seven years later [14].

7.9. Type 2: psychiatric disorders

The only description of psychiatric treatment is that of Ganjavi et al., in a girl with type 2, an intelligence quotient of 70–75, severe anxiety, obsessions, self-harm and a suicide attempt by drug overdose at 17 years of age [71]. Selective serotonin reuptake inhibitors (escitalopram, fluoxetine, fluvoxamine and sertraline) appeared to increase agitation; the antipsychotics risperidone, paliperidone, quetiapine and chlorpromazine increased agitation or caused excessive sedation, with rebound as the effect wore off; benzodiazepines (clonazepam, lorazepam) helped sleep without improving mood or behaviour [71]. During admission, gabapentin, started at 100 mg three times daily and increased to 200 mg three times daily, had a calming effect and reduced agitation and self-harm; aripiprazole, started at 5 mg in the morning and gradually increased to 15 mg per day, improved anxiety, mood and frustration tolerance within days, self-harm ceased, and worsening on stopping or reducing the dose reversed on reintroduction [71]. The authors suggest considering aripiprazole for mood and behavioural disorders in type 2 or other developmental delay and gabapentin for agitation, acknowledging that it is not known whether the response is related to the deletion [71]. This is a single case and off-label use for gabapentin; aripiprazole has a US indication for irritability associated with autism, an analogous condition whose dosing serves as a reference [101].

The US Abilify® label recommends, for irritability associated with autistic disorder between 6 and 17 years of age, starting at 2 mg/day, increasing to 5 mg/day and, if necessary, to 10 or 15 mg/day, with adjustments of up to 5 mg/day at intervals of at least one week, within a therapeutic range of 5–15 mg/day; in cytochrome P450 2D6 (CYP2D6) poor metabolisers half the dose is used [101]. In the European Union, irritability associated with autism is not an approved indication under 18 years of age, and the summary of product characteristics gives no dosing; the European paediatric indications are schizophrenia from 15 years of age (starting with 2 mg for 2 days, 5 mg for 2 days and a target dose of 10 mg/day, with increments of 5 mg up to a maximum of 30 mg/day) and manic episodes of bipolar I disorder from 13 years of age, for up to 12 weeks [102]. The most common adverse effects in paediatric patients with autism were sedation, fatigue, vomiting, somnolence, tremor, pyrexia, drooling, decreased appetite, extrapyramidal disorder and lethargy; warnings include neuroleptic malignant syndrome (discontinue and monitor), tardive dyskinesia, metabolic changes (hyperglycaemia, dyslipidaemia and weight gain, with monitoring of blood glucose and weight), orthostatic hypotension, leucopenia and neutropenia (frequent full blood counts in the first months in patients with a history), seizures, compulsive behaviours and suicidal ideation [101].

The US Neurontin® label has no psychiatric indication; in epilepsy from 12 years of age, the starting dose is 300 mg three times daily, titratable up to 600 mg three times daily, and between 3 and 11 years treatment starts at 10–15 mg/kg/day in three doses, titrated over about 3 days to 40 mg/kg/day (3–4 years) or 25–35 mg/kg/day (5–11 years), with adjustment for renal function [103]. The dose used by Ganjavi et al. was below the labelled antiepileptic dose [71, 103]. Warnings include drug reaction with eosinophilia and systemic symptoms (DRESS), anaphylaxis and angioedema (discontinue), somnolence and dizziness, suicidal ideation, seizures after abrupt withdrawal, respiratory depression with central nervous system depressants such as benzodiazepines and opioids, and neuropsychiatric reactions, including hostility, in children aged 3 to 12 years [103]. The combination with benzodiazepines, used in the case described, requires respiratory monitoring [71, 103].

7.10. MPAPA and 2p24 duplication: tumours and therapeutic precautions

The neuroblastomas of the two patients with MPAPA were treated with chemotherapy and surgery according to usual risk stratification, and both were in remission [12]. The tumour surveillance recommended by extrapolation is described in 11.2. Because the somatropin label contraindicates its use in active neoplasia and requires monitoring of pre-existing tumours, any prescription of GH to a patient with MPAPA or 2p24 duplication, both with predisposition to neuroblastoma, requires this consideration; the sources do not describe such use [23, 99].

7.11. Drugs with published data in the entities in scope

7.12. Comparative regulatory status

The sources consulted do not include the INFARMED or Brazilian Health Regulatory Agency (ANVISA) summaries for these drugs.

7.13. Intervention evidence

There are no registered clinical trials for any of the entities: a ClinicalTrials.gov search on 26 October 2026 with the terms "Feingold syndrome", "MIR17HG", "oculodigitoesophagoduodenal" and "megalencephaly polydactyly MYCN" returned zero studies, and GeneReviews acknowledges that there may be no trials [3, 104]. The available evidence is as follows.

Data from patients with the syndrome come from single cases, without a comparison group, and do not allow efficacy to be estimated; the antacid prophylaxis studies are observational and were cited by the ERNICA consensus, not read in full text; NSIGHT2 included only part of the population of interest [22, 94]. No published study has evaluated neuroblastoma surveillance in MPAPA [12, 23].

7.14. Developments in the last 6–12 months

Between October 2025 and October 2026 there were no clinical trials or regulatory procedures, and the developments are of a low level of evidence. Ferroul et al. (December 2025) described three patients with type 1 without atresia and with laryngeal cleft, congenital deafness, agenesis of the corpus callosum and radioulnar synostosis, and proposed major criteria and assessment recommendations (case series) [59]. Samara et al. (January 2026) reported the prenatal diagnosis of a 342 kb deletion in a pregnancy complicated by severe pre-eclampsia and fetal growth restriction (single case, no causal link demonstrated) [15]. Mérida De la Torre et al. (May 2026) described the first segregating N-terminal missense variant, classified as a VUS (family report) [46]. Segal (2026) published an interview with the mother of monozygotic twins with the syndrome, without clinical data in the abstract [105]. ClinGen classified the relationship of MYCN with megalencephaly-polydactyly syndrome as moderate on 7 January 2026, and ClinVar re-evaluated p.(Thr58Met) as pathogenic/likely pathogenic on 12 August 2026 (curation) [7, 50]. The 2024 review by Nishio et al. cites no patients with MPAPA other than the three described, and no later reports were located up to October 2026, which does not prove that none exist [19].

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

The sources identify no drugs contraindicated specifically because of the syndrome; the precautions are those of each drug and of surgery [3]. Somatropin should not be used with active neoplasia [99]. In pregnancy, data on affected women are scarce: in the family of Klaniewska et al., the carrier mother had three affected children, two with oesophageal atresia, deliveries at 38 and 42 weeks and polyhydramnios in the second pregnancy [29]; in the other prenatal reports, the carrier was the father [15, 18]. The severe pre-eclampsia described in a pregnancy with an affected fetus is an isolated finding [15]. The labels mention, for aripiprazole, a risk of extrapyramidal and withdrawal symptoms in the newborn exposed in the third trimester; for gabapentin and omeprazole, epidemiological data have not shown an increase in major malformations, with methodological limitations [97, 101, 103]. The sources consulted do not describe lactation in these patients.

In the perioperative period, oesophageal dilatations are performed under general anaesthesia with tracheal intubation [77], and the ERNICA consensus provides for preoperative tracheoscopy, a central venous catheter and arterial line, perioperative antibiotics and pressure and duration limits for thoracoscopy [76]. In life-threatening events, multidisciplinary assessment precedes any surgery [77].

Dysphagia and reflux persist into adulthood, with more oesophagitis and Barrett metaplasia in adults operated on for oesophageal atresia, without a demonstrated increase in oesophageal cancer; ESPGHAN-NASPGHAN recommends transition to an adult physician with experience of the condition, regular clinical follow-up (dysphagia, reflux, respiratory symptoms, anaemia) and endoscopy with biopsies at transition and every 5–10 years [77, 94]. Only 55% of ERNICA and affiliated centres had a transition pathway, started at a median of 18 years, and only two centres (7%) had a transition education programme [95].

7.16. Applicable guidelines and divergences

8. Reproductive Options and Family Counselling

8.1. Inheritance and recurrence risks

All entities of the class have autosomal dominant transmission. In type 1, about 60% of those diagnosed have an affected parent and about 40% have a de novo variant, according to Marcelis et al. and personal communication from the GeneReviews authors; the summary of the same document states that the de novo proportion is unknown, an internal inconsistency, and the figure of about 40% is the one supported by the body of the text [3, 107]. In type 2 there is documented vertical transmission, from father to son, and deletions that segregate in families, as well as presumed de novo cases [13, 14]. The three patients with MPAPA had a de novo variant [12].

The comparison with non-syndromic atresias is central to counselling: the recurrence risk of oesophageal atresia in a family without a syndrome is 0.2–2%, whereas 30–40% of patients with Feingold syndrome have oesophageal atresia [29]. Because atresia occurs in only some carriers, the risk of an affected individual's child having atresia is less than 50%; the series do not allow this figure to be calculated precisely for each family [3, 29].

8.2. Parental and cascade testing

The parents of a proband with an apparently de novo variant should be tested, and the assessment should include clinical examination and hand radiographs, because relatives with minimal expression, such as isolated microcephaly, are recognised only after testing [3, 9, 16]. An apparently negative family history is confirmed only after this assessment [3]. The parent in whom the variant arose may have somatic mosaicism and be only mildly affected [3]. The risk to other relatives depends on the status of the parents [3]. In the two recent prenatal reports, the fetal deletion led to the diagnosis of mildly affected fathers, and in one of them also of the paternal grandmother [15, 18]. The molecular diagnosis allows screening of at-risk relatives and the offer of preimplantation genetic testing [17]. When investigating an apparent de novo variant, alternative paternity or maternity and undisclosed adoption should be considered [3].

8.3. Prenatal and preimplantation diagnosis

Once the familial variant has been identified, prenatal diagnosis and preimplantation genetic testing are possible [3]. In any pregnancy with a known carrier fetus or a 50% risk, GeneReviews recommends high-resolution ultrasound with fetal echocardiography, given the risk of gastrointestinal, cardiac and renal malformations [3]. Detection of an atresia requiring early surgery determines the place and planning of delivery [15, 94]. Perspectives on the use of prenatal diagnosis for termination of pregnancy vary among families and professionals [3]. In the Portuguese case, multidisciplinary counselling supported the couple, who decided to continue the pregnancy [18].

8.4. Counselling content and minors

Counselling should begin before pregnancy and be offered to affected or at-risk young adults [3]. Variable expressivity within the same family makes it impossible to predict severity in a child from a parent's phenotype [29, 57]. Information on cognitive prognosis, generally favourable in type 1, should avoid both overestimation and underestimation [3, 9]. DNA banking should be offered when the molecular diagnosis is not established [3, 80]. In type 1, clarifying the status of an at-risk newborn or infant guides the renal, cardiac, audiological and developmental assessment recommended after diagnosis [3]. In MPAPA and 2p24 duplication, the AACR recommends early predictive testing in at-risk infants, because neuroblastoma may be present at birth [23]. Klaniewska et al. recommend genetic counselling for all patients with oesophageal atresia and their families [29].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

There is no validated estimate of prevalence or incidence for any of the entities of the class [3]. Orphanet places types 1 and 2 in the point prevalence class below 1/1,000,000, with about 120 cases for type 1 from a specialist source; this is the upper limit of a class and not a measured rate [9, 11]. Complete penetrance of the digital features, the frequency of mildly affected parents and the identification of deletions by CMA, often classified in other categories, suggest under-detection [9]. The figures available are counts of published cases: 116 affected individuals from 69 families in the 2019 GeneReviews review, more than 120 patients with type 1 by 2021 and more than 200 cases according to a 2026 count with no primary source given [3, 15, 16]; 23 patients with type 2 by 2022 [17]; and 3 patients with MPAPA by 2024 [19].

The epidemiological context of the associated malformations helps to interpret the frequency of the syndrome. Gastrointestinal obstruction of all types occurs in about 1 in 2,000 births, tracheoesophageal fistula in about 1 in 3,000 live births, with a slight male predominance and association with oesophageal atresia in more than 85% of cases, and the VACTERL association in 1.6 per 10,000 [20, 75]. About half of patients with tracheoesophageal fistula have an associated major malformation, and only about 45% are isolated anomalies [20]. Feingold syndrome has been described as the most frequent cause of familial syndromic gastrointestinal atresia [38], but in a consecutive series of 148 cases of oesophageal atresia from Paris only 6% had a known genomic cause [64]. Trisomy 21 remains the most frequent cause of duodenal atresia (30–50%) [20].

9.2. Geography, ancestry and founder effect

The series and reports come mainly from Europe: the Netherlands (the three-generation family with 11 affected individuals and the Nijmegen series) [34, 38], France, including Réunion Island [28, 59], Italy [16], Belgium [32], Poland [29], Hungary [69], Spain [46], Greece [15], Lithuania [41] and the United Kingdom [62, 66, 67]. Outside Europe there are reports from the USA [5, 14, 70], Canada [71, 74], Australia [63], Turkey [17, 61, 107], Kosovo [65], Israel [51], Japan [12, 40, 52], China [58] and Taiwan [21]. One of the North American patients with type 2 was the son of a mother of Chinese ancestry and a European father [14]. No founder variant or population clustering has been described; the only founder population mentioned in the differential is the French-Canadian one, in multiple intestinal atresias due to TTC7A [20]. The geographical distribution of reports reflects where there are genetics services and publication, not the real frequency of the syndrome.

In Portugal, Gouveia et al. published in 2023, from Vila Nova de Gaia and Porto, the prenatal diagnosis of a 1.5 Mb deletion at 2p24.3 including MYCN in a fetus with isolated microcephaly, inherited from the father [18]. From Brazil, the sources consulted include no published cases; the Brazilian textbook of genetic syndromes includes Feingold syndrome in the differential diagnosis of the VACTERL association [83]. The absence of Brazilian data in the sources does not demonstrate that no cases exist.

9.3. Special populations

Variability between monozygotic twins was described by Blaumeiser et al., and in 2026 an interview with the mother of monozygotic twins with the syndrome was published [57, 105]. Data on adults are scarce and concentrated on type 2: a 28-year-old man with chronic musculoskeletal pain, contractures, Raynaud phenomenon, depression and sleep disturbances, his 70-year-old father with obstructive sleep apnoea, late-onset hypogonadotropic hypogonadism, cataracts and coronary disease, working as an accountant, and a 58-year-old woman with keratoconus [14, 55]. None of these findings has an established causal link with the deletion [14]. In type 1, most adults live independently, and several were diagnosed only as parents of affected children, with occupations such as farmer and tractor driver [3, 15, 29]. The sources do not describe pregnancy outcomes in women with type 2 or pregnancy series in type 1 [29].

9.4. Reference networks and patient associations

The relevant European networks are ERN ITHACA, for congenital malformations and intellectual disability, and ERNICA, for atresias and other digestive anomalies [24, 25]. EAT (Federation of Esophageal Atresia and Tracheo-Esophageal Fistula support groups) federates the European oesophageal atresia associations, TOFS (Tracheo-Oesophageal Fistula Support) supports families in the United Kingdom, and Unique supports families with rare chromosomal disorders, including 2p24 and 13q deletions [109–111]. GeneReviews also lists the Children's Craniofacial Association, Face Equality International, the Learning Disabilities Association of America, MedlinePlus and REACH, a British association for upper limb differences [3]. The sources consulted do not identify specific Portuguese or Brazilian associations.

10. Illustrative Clinical Case

10.1. Prenatal diagnosis of Feingold syndrome type 1 from isolated fetal microcephaly (Portugal)

Gouveia et al. described a woman aged 30–39 years, G3P1, with no relevant history and a negative first-trimester combined screening [18]. At the anomaly scan at 21+2 weeks, the head circumference was 1 standard deviation below the mean (172.2 mm), and the examination was incomplete owing to suboptimal imaging; on repetition at 23+6 weeks, the head circumference had decelerated to more than 2 standard deviations below the mean (191.5 mm, below the 5th centile), with the other biometric measurements on the 5th centile and no other malformations [18]. A team of obstetricians, a paediatrician and a geneticist proposed amniocentesis, performed at 24+1 weeks: cytomegalovirus testing and QF-PCR for chromosomes 13, 18, 21, X and Y were negative, and fetal aCGH (Agilent 8×60k) showed a deletion of about 1.5 Mb at 2p24.3 including MYCN [18]. Fetal MRI at 24+0 weeks confirmed the microcephaly, with supratentorial and infratentorial biometry more than 2 weeks below the mean, and fetal echocardiography was normal [18]. The father had the same deletion, 2–3 toe syndactyly, mild learning difficulty and a head circumference of 55 cm (3rd centile); the mother was normal, and the paternal grandmother was later found to be affected as well [18].

The couple decided to continue the pregnancy [18]. At 33+1 weeks, the microcephaly persisted (2–3 standard deviations below the mean), with no other malformations; delivery was spontaneous vaginal at 39+6 weeks, and the newborn girl weighed 2,850 g, with an Apgar score of 9/10, microcephaly, bilateral clinodactyly of the 5th finger, bilateral shortening of the middle phalanges of the 2nd and 5th fingers, bilateral 2–3 toe syndactyly and short palpebral fissures [18]. Echocardiography, electrocardiography, abdominal ultrasound (to exclude atresia), liver and renal function and transfontanellar ultrasound were normal, as in the father [18]. At 13 months, height and weight were between the 3rd and 15th centiles, head circumference slightly below the 3rd centile and neuromotor development without major abnormalities [18]. The photographs of the feet with syndactyly (Figure 6) and of the hands with clinodactyly of the 5th finger (Figure 7) show the signs that would have allowed the father to be recognised before the fetal diagnosis; the original legend of Figure 6 refers to "syndactyly of the fifth finger", but the image is of the feet and the syndactyly described in the text is of toes 2–3 [18]. The article places the first report of the syndrome in 1997, whereas the original description dates from 1975 [18, 26].

The case illustrates three points of this article: fetal microcephaly may be the only ultrasound sign, the deletion may be inherited from a parent with mild expression who was never diagnosed, and targeted neonatal assessment (echocardiography, abdominal and renal ultrasound) follows the general recommendations [3, 18]. Amniocentesis at 24 weeks was later than the ideal window of 15–17 weeks proposed by the authors themselves, because the microcephaly only became evident at the second examination [18].

10.2. Complementary case: oesophageal atresia in siblings from a three-generation family (Poland)

Klaniewska et al. described two brothers with type C oesophageal atresia, both operated on thoracoscopically on day 2 of life, with anastomotic stricture treated by endoscopic dilatations (at 13 and 15 months in one, at 14, 20 and 23 months in the other) and, in the second, food bolus impaction at 27 months without stricture [29]. The exome identified NM_005378.6:c.266dup p.(Ser90GlnfsTer176), predicted to undergo NMD, also present in a 9-year-old brother without atresia and with intellectual disability, in the mother, who had only microcephaly, and in the maternal grandfather, with microcephaly and borderline intellect [29]. A cousin operated on for duodenal atresia did not have the variant, showing that not every atresia in an affected family belongs to the syndrome [29]. The pedigree and sequencing (Figure 1), the phenotypes (Figure 2) and the toe and finger syndactyly (Figure 3) are in the article; the legends were consulted [29]. The dilatations followed the approach to symptomatic strictures advocated by ESPGHAN-NASPGHAN [77].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

In type 1, gastrointestinal atresia is the main determinant of early morbidity; after repair, reports describe treatable anastomotic strictures and normal or near-normal feeding by preschool age [3, 29]. Most adults live independently, and overt intellectual disability is rare [3]. Vesicoureteral reflux with renal dysplasia can lead to renal failure, including end-stage kidney disease at 6 years of age in one report [3, 67]. Profound hearing loss with cochlear nerve agenesis limits rehabilitation [68]. Larger 2p24 deletions are associated with more marked intellectual disability [5, 15]. In type 2 there are adults in the seventh and eighth decades with a working life, and the morbidities described are short stature, aortic dilatation, memory, sleep and mood problems and chronic musculoskeletal pain [14, 55]. In MPAPA, the first patient had severe intellectual disability, and the two neuroblastomas went into remission after treatment [12]. The sources have no survival or life expectancy data for any of the entities.

11.2. Surveillance

In type 1 and type 2 there is no oncological surveillance, because loss of function of MYCN and deletion of MIR17HG confer no known tumour risk [3, 9]. The neuroblastoma protocol applies to MPAPA by extrapolation: to recommend surveillance, the AACR requires segregation with neuroblastoma in at least two individuals from two pedigrees or a neuroblastoma prevalence of at least 1% in carriers, and it included 2p24 duplication, also a gain of MYCN dosage, but not MPAPA, with neuroblastoma in 2 of the 3 patients described [12, 23]. Nishio et al. advocate treating MPAPA as a hereditary cancer predisposition syndrome, without specifying a protocol [12]. Because 80% of neuroblastomas are diagnosed by 6 years of age and 98% by 10 years, surveillance ends at 10 years; MRI replaces ultrasound when the latter is technically inadequate [23, 106]. The 2024 version reduced the frequency of chest radiographs compared with the 2017 version, which performed them at every visit [23, 106]. The 2025 synthesis of the same workshop identified nearly 100 paediatric cancer predisposition syndromes that benefit from targeted surveillance and noted that few studies have evaluated the effectiveness of this surveillance [112].

11.3. Transition to adulthood

Patients operated on for oesophageal atresia should move to an adult physician with experience of the condition, with a specific transition programme, endoscopy at transition and every 5–10 years, and Barrett surveillance according to the consensus statements [77, 94, 96]. Educational and social transition begins in adolescence [3]. In type 2, follow-up echocardiography and attention to mood, memory, sleep and musculoskeletal pain continue into adulthood [14]. In patients with type 2 treated with GH, the PES recommends reassessing the somatotropic axis at transition, after at least one month off GH, in isolated idiopathic deficiency or with a small pituitary, starting with IGF-1 and moving to stimulation testing if it is low, and waives retesting in deficiency with three or more pituitary deficiencies, a documented causal variant or a specific structural hypothalamic-pituitary defect (except ectopic neurohypophysis) [98]. The guideline does not address MIR17HG deletion or Rathke's cleft cyst, so classifying a patient such as that of Muriello et al. into one of these groups is an individual decision [14, 98].

12. Summary and Final Considerations

Feingold syndrome is a class of autosomal dominant disorders defined by reduced dosage of a developmental axis in which MYCN activates the miR-17~92 cluster. Type 1 results from MYCN haploinsufficiency, due to loss-of-function variants or deletions that may encompass contiguous genes at 2p24, and combines brachymesophalangy of the 2nd and 5th fingers, toe syndactyly, microcephaly, short palpebral fissures and, in about one third to one half of patients, oesophageal or duodenal atresia, with generally mild learning disability [3, 38]. Type 2 results from deletion of MIR17HG at 13q31.3, shares the skeletal pattern and microcephaly, has no atresia and adds more frequent short stature, growth hormone deficiency, bicuspid aortic valve and aortic dilatation [14, 17]. MYCN gain of function produces the inverse picture, with megalencephaly, postaxial polydactyly, ventriculomegaly and neuroblastoma in infancy [12].

Recognition depends on examining the hands and feet of every newborn with atresia and on assessing the parents, because about 60% of type 1 cases are inherited from parents whose expression is sometimes minimal [3, 16]. CMA detects the deletions of both types, and MYCN sequencing with copy-number analysis or exome sequencing detects point variants; about one third of patients with a clinical diagnosis remain without an identified cause, and severe phenotypes justify searching for a second diagnosis [3, 16, 28]. Variants should be reported on transcript NM_005378.6, and missense variants outside the bHLH domain require additional evidence before being considered pathogenic [10, 46].

There is no specific treatment. Management follows the ESPGHAN-NASPGHAN and ERNICA oesophageal atresia guidelines, with surgery at specialist centres, acid suppression, endoscopic surveillance throughout childhood and adulthood and structured transition, together with renal, cardiac, audiological and developmental surveillance [3, 76, 77, 94]. In type 2, echocardiography at diagnosis and investigation of growth hormone deficiency in the presence of short stature are recommendations of the authors of the largest series [14]. In MPAPA, neuroblastoma surveillance until 10 years of age applies by extrapolation from the AACR protocol [23]. The recurrence risk is 50% for the children of a carrier, much higher than that of non-syndromic atresias and of the VACTERL association, which makes the differential diagnosis decisive in counselling [3, 29, 83].

13. References

1. Orphanet. Feingold syndrome (ORPHA:1305). Available from: https://www.orpha.net/en/disease/detail/1305
2. Orphanet. Feingold syndrome type 1 (ORPHA:391641). Available from: https://www.orpha.net/en/disease/detail/391641
3. Marcelis CLM, de Brouwer APM. Feingold Syndrome 1. 2009 Jun 30 [Updated 2019 Apr 4]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK7050/
4. Orphanet. Feingold syndrome type 2 (ORPHA:391646). Available from: https://www.orpha.net/en/disease/detail/391646
5. Burnside RD, Molinari S, Botti C, Brooks SS, Chung WK, Mehta L, et al. Features of Feingold syndrome 1 dominate in subjects with 2p deletions including MYCN. Am J Med Genet A. 2018;176(9):1956-1963. doi:10.1002/ajmg.a.40355
6. Online Mendelian Inheritance in Man, OMIM®. #620748 Megalencephaly-polydactyly syndrome; MPAPA. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/620748
7. ClinGen. Gene-disease validity: MYCN — Feingold syndrome type 1 (Definitive, 2025-07-02) and megalencephaly-polydactyly syndrome (Moderate, 2026-01-07); Dosage sensitivity: MYCN (HI 3, 2017-07-12); region ISCA-46287 2p24.3 MYCN-DDX1 duplication (TS 2). Available from: https://search.clinicalgenome.org/kb/genes/HGNC:7559
8. Orphanet. Megalencephaly-postaxial polydactyly-ventriculomegaly-neuroblastoma syndrome (ORPHA:730286). Available from: https://www.orpha.net/en/disease/detail/730286
9. GeneGuidance (user database). Record for the MYCN gene and linked conditions (OMIM 164280 and 620748): identifiers, clinical curation, Orphadata frequencies, ClinGen and listed references. Export of 26 Sep 2026, accessed 26 Oct 2026.
10. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348
11. GeneGuidance (user database). Record for the MIR17HG gene and the OMIM 614326 condition: identifiers, clinical curation, Orphadata frequencies and listed references. Export of 26 Sep 2026, accessed 26 Oct 2026.
12. Nishio Y, Kato K, Tran Mau-Them F, Futagawa H, Quélin C, Masuda S, et al. Gain-of-function MYCN causes a megalencephaly-polydactyly syndrome manifesting mirror phenotypes of Feingold syndrome. HGG Adv. 2023;4(4):100238. doi:10.1016/j.xhgg.2023.100238
13. de Pontual L, Yao E, Callier P, Faivre L, Drouin V, Cariou S, et al. Germline deletion of the miR-17∼92 cluster causes skeletal and growth defects in humans. Nat Genet. 2011;43(10):1026-30. doi:10.1038/ng.915
14. Muriello M, Kim AY, Sondergaard Schatz K, Beck N, Gunay-Aygun M, Hoover-Fong JE. Growth hormone deficiency, aortic dilation, and neurocognitive issues in Feingold syndrome 2. Am J Med Genet A. 2019;179(3):410-416. doi:10.1002/ajmg.a.61037
15. Samara AA, Perros P, Koutras A, Janho MB, Manolakos E, Daponte N, et al. Prenatal Diagnosis of a Feingold Syndrome Pregnancy Complicated with Severe Preeclampsia: A Report of a Challenging Case. Genes (Basel). 2026;17(1):. doi:10.3390/genes17010054
16. Tedesco MG, Lonardo F, Ceccarini C, Cesarano C, Digilio MC, Magliozzi M, et al. Clinical and molecular characterizations of 11 new patients with type 1 Feingold syndrome: Proposal for selecting diagnostic criteria and further genetic testing in patients with severe phenotype. Am J Med Genet A. 2021;185(4):1204-1210. doi:10.1002/ajmg.a.62068
17. Demir Ş, Söylemez MA, Arman A, Ata P. First Patient Diagnosed as Feingold Syndrome Type 2 with Alport Syndrome and Review of the Current Literature. Mol Syndromol. 2022;13(5):447-453. doi:10.1159/000524058
18. Gouveia I, Geraldo AF, Godinho C, Castedo S. Feingold syndrome type 1: a rare cause of fetal microcephaly (prenatal diagnosis). BMJ Case Rep. 2023;16(3):. doi:10.1136/bcr-2022-254366
19. Nishio Y, Kato K, Oishi H, Takahashi Y, Saitoh S. MYCN in human development and diseases. Front Oncol. 2024;14:1417607. doi:10.3389/fonc.2024.1417607
20. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Oesophageal and intestinal atresia, p. 266–269.
21. Chen CP, Lin SP, Chern SR, Wu PS, Chang SD, Ng SH, et al. A de novo 4.4-Mb microdeletion in 2p24.3 → p24.2 in a girl with bilateral hearing impairment, microcephaly, digit abnormalities and Feingold syndrome. Eur J Med Genet. 2012;55(11):666-9. doi:10.1016/j.ejmg.2012.07.003
22. Manickam K, McClain MR, Demmer LA, Biswas S, Kearney HM, Malinowski J, et al. Exome and genome sequencing for pediatric patients with congenital anomalies or intellectual disability: an evidence-based clinical guideline of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2021;23(11):2029-2037. doi:10.1038/s41436-021-01242-6
23. Kamihara J, Diller LR, Foulkes WD, Michaeli O, Nakano Y, Pajtler KW, et al. Neuroblastoma Predisposition and Surveillance-An Update from the 2023 AACR Childhood Cancer Predisposition Workshop. Clin Cancer Res. 2024;30(15):3137-3143. doi:10.1158/1078-0432.CCR-24-0237
24. ERN ITHACA — European Reference Network on Rare Congenital Malformations and Rare Intellectual Disability. Available from: https://ern-ithaca.eu/
25. ERNICA — European Reference Network on Rare Inherited and Congenital (digestive and gastrointestinal) Anomalies. Available from: https://www.ern-ernica.eu/
26. Feingold M. Case Report 30. Synd Ident. 1975;3:16–7.
27. Online Mendelian Inheritance in Man, OMIM®. #164280 Feingold syndrome 1; FGLDS1. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/164280
28. Cognet M, Nougayrede A, Malan V, Callier P, Cretolle C, Faivre L, et al. Dissection of the MYCN locus in Feingold syndrome and isolated oesophageal atresia. Eur J Hum Genet. 2011;19(5):602-6. doi:10.1038/ejhg.2010.225
29. Klaniewska M, Toczewski K, Rozensztrauch A, Bloch M, Dzielendziak A, Gasperowicz P, et al. Occurrence of Esophageal Atresia With Tracheoesophageal Fistula in Siblings From Three-Generation Family Affected by Variable Expressivity MYCN Mutation: A Case Report. Front Pediatr. 2021;9:783553. doi:10.3389/fped.2021.783553
30. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 110.
31. Brunner HG, Winter RM. Autosomal dominant inheritance of abnormalities of the hands and feet with short palpebral fissures, variable microcephaly with learning disability, and oesophageal/duodenal atresia. J Med Genet. 1991;28(6):389-94. doi:10.1136/jmg.28.6.389
32. Courtens W, Levi S, Verbelen F, Verloes A, Vamos E. Feingold syndrome: report of a new family and review. Am J Med Genet. 1997;73(1):55-60.
33. Kawame H, Pagon RA, Hudgins L. Digital anomalies, microcephaly, and normal intelligence: new syndrome or Feingold syndrome? Am J Med Genet. 1997;69(3):240-4. doi:10.1002/(sici)1096-8628(19970331)69:3<240::aid-ajmg6>3.0.co;2-p
34. Celli J, van Beusekom E, Hennekam RC, Gallardo ME, Smeets DF, de Córdoba SR, et al. Familial syndromic esophageal atresia maps to 2p23-p24. Am J Hum Genet. 2000;66(2):436-44. doi:10.1086/302779
35. Celli J, van Bokhoven H, Brunner HG. Feingold syndrome: clinical review and genetic mapping. Am J Med Genet A. 2003;122A(4):294-300. doi:10.1002/ajmg.a.20471
36. Online Mendelian Inheritance in Man, OMIM®. *164840 MYCN proto-oncogene, bHLH transcription factor; MYCN. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/164840
37. van Bokhoven H, Celli J, van Reeuwijk J, Rinne T, Glaudemans B, van Beusekom E, et al. MYCN haploinsufficiency is associated with reduced brain size and intestinal atresias in Feingold syndrome. Nat Genet. 2005;37(5):465-7. doi:10.1038/ng1546
38. Marcelis CL, Hol FA, Graham GE, Rieu PN, Kellermayer R, Meijer RP, et al. Genotype-phenotype correlations in MYCN-related Feingold syndrome. Hum Mutat. 2008;29(9):1125-32. doi:10.1002/humu.20750
39. Tassano E, Di Rocco M, Signa S, Gimelli G. De novo 13q31.1-q32.1 interstitial deletion encompassing the miR-17-92 cluster in a patient with Feingold syndrome-2. Am J Med Genet A. 2013;161A(4):894-6. doi:10.1002/ajmg.a.35781
40. Kato K, Miya F, Hamada N, Negishi Y, Narumi-Kishimoto Y, Ozawa H, et al. MYCN de novo gain-of-function mutation in a patient with a novel megalencephaly syndrome. J Med Genet. 2019;56(6):388-395. doi:10.1136/jmedgenet-2018-105487
41. Siavrienė E, Preikšaitienė E, Maldžienė Ž, Mikštienė V, Rančelis T, Ambrozaitytė L, et al. A de novo 13q31.3 microduplication encompassing the miR-17 ~ 92 cluster results in features mirroring those associated with Feingold syndrome 2. Gene. 2020;753:144816. doi:10.1016/j.gene.2020.144816
42. Hempel A, Pagnamenta AT, Blyth M, Mansour S, McConnell V, Kou I, et al. Deletions and de novo mutations of SOX11 are associated with a neurodevelopmental disorder with features of Coffin-Siris syndrome. J Med Genet. 2016;53(3):152-62. doi:10.1136/jmedgenet-2015-103393
43. Kee et al. Genomics in Skin, Neurological and Rare Cancers. In: ESMO Handbook — Cancer Genomics. Lugano: ESMO Press. Neuroblastoma, p. 102.
44. Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. Retinoblastoma, p. 174.
45. Jopling CL, Willis AE. N-myc translation is initiated via an internal ribosome entry segment that displays enhanced activity in neuronal cells. Oncogene. 2001;20(21):2664-70. doi:10.1038/sj.onc.1204404
46. Javier Mérida De la Torre F, Porta Pelayo J, Ortiz-Martín I. Expanding the MYCN Variant Spectrum in Feingold Syndrome Type 1: A Novel N-Terminal Missense Variant Segregating in an Affected Family. Genes (Basel). 2026;17(5):. doi:10.3390/genes17050552
47. Knoepfler PS, Cheng PF, Eisenman RN. N-myc is essential during neurogenesis for the rapid expansion of progenitor cell populations and the inhibition of neuronal differentiation. Genes Dev. 2002;16(20):2699-712. doi:10.1101/gad.1021202
48. Ota S, Zhou ZQ, Keene DR, Knoepfler P, Hurlin PJ. Activities of N-Myc in the developing limb link control of skeletal size with digit separation. Development. 2007;134(8):1583-92. doi:10.1242/dev.000703
49. Charron J, Gagnon JF, Cadrin-Girard JF. Identification of N-myc regulatory regions involved in embryonic expression. Pediatr Res. 2002;51(1):48-56. doi:10.1203/00006450-200201000-00010
50. National Center for Biotechnology Information. ClinVar: MYCN variants VCV000433151, VCV000545970, VCV000013892, VCV000013894, VCV002443779, VCV002445219. Available from: https://www.ncbi.nlm.nih.gov/clinvar/
51. Peleg A, Kurolap A, Sagi-Dain L, Larom-Khan G, Adir V, Mory A, et al. A novel mutation in MYCN gene causing congenital absence of the flexor pollicis longus tendon as an unusual presentation of Feingold syndrome 1. Clin Dysmorphol. 2021;30(2):71-75. doi:10.1097/MCD.0000000000000342
52. Nitta Y, Sano H, Yoshimura Y, Takada F, Yamashita T. Feingold syndrome with GJB2 variants. Auris Nasus Larynx. 2025;52(4):522-526. doi:10.1016/j.anl.2025.07.005
53. Online Mendelian Inheritance in Man, OMIM®. #614326 Feingold syndrome 2; FGLDS2. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/614326
54. Fiori E, Babicola L, Andolina D, Coassin A, Pascucci T, Patella L, et al. Neurobehavioral Alterations in a Genetic Murine Model of Feingold Syndrome 2. Behav Genet. 2015;45(5):547-59. doi:10.1007/s10519-015-9724-8
55. Sirchia F, Di Gregorio E, Restagno G, Grosso E, Pappi P, Talarico F, et al. A case of Feingold type 2 syndrome associated with keratoconus refines keratoconus type 7 locus on chromosome 13q. Eur J Med Genet. 2017;60(4):224-227. doi:10.1016/j.ejmg.2017.01.010
56. Mirzamohammadi F, Kozlova A, Papaioannou G, Paltrinieri E, Ayturk UM, Kobayashi T. Distinct molecular pathways mediate Mycn and Myc-regulated miR-17-92 microRNA action in Feingold syndrome mouse models. Nat Commun. 2018;9(1):1352. doi:10.1038/s41467-018-03788-7
57. Blaumeiser B, Oehl-Jaschkowitz B, Borozdin W, Kohlhase J. Feingold syndrome associated with two novel MYCN mutations in sporadic and familial cases including monozygotic twins. Am J Med Genet A. 2008;146A(17):2304-7. doi:10.1002/ajmg.a.32444
58. Lei J, Han L, Huang Y, Long M, Zhao G, Yan S, et al. Feingold syndrome type 2 in a patient from China. Am J Med Genet A. 2021;185(7):2262-2266. doi:10.1002/ajmg.a.62190
59. Ferroul F, Snanoudj S, Leterme G, Mezouaghi K, Kieffer-Traversier M, Celse T, et al. Rare features in Feingold syndrome type 1. Eur J Med Genet. 2025;78:105049. doi:10.1016/j.ejmg.2025.105049
60. Ruiz-Pérez MV, Henley AB, Arsenian-Henriksson M. The MYCN Protein in Health and Disease. Genes (Basel). 2017;8(4):. doi:10.3390/genes8040113
61. Koçak H, Ozaydin E, Köse G, Marcelis CL, Kamsteeg EJ, Ceylaner S. A Feingold syndrome case with previously undescribed features and a new mutation. Genet Couns. 2009;20(3):261-7.
62. Holder-Espinasse M, Ahmad Z, Hamill J, Pahari A, Misra D, Drake D, et al. Familial syndromic duodenal atresia: Feingold syndrome. Eur J Pediatr Surg. 2004;14(2):112-6. doi:10.1055/s-2004-815858
63. Büttiker V, Wojtulewicz J, Wilson M. Imperforate anus in Feingold syndrome. Am J Med Genet. 2000;92(3):166-9. doi:10.1002/(sici)1096-8628(20000529)92:3<166::aid-ajmg2>3.0.co;2-#
64. Ranza E, Le Gouez M, Guimier A, Dunlop NK, Beaudoin S, Malan V, et al. Retrospective evaluation of clinical and molecular data of 148 cases of esophageal atresia. Am J Med Genet A. 2023;191(1):77-83. doi:10.1002/ajmg.a.62989
65. Zeka N, Bejiqi R, Gerguri A, Zogaj L, Jashari H. A new variant of MYCN gene as a cause of Feingold syndrome. Clin Case Rep. 2022;10(5):e05886. doi:10.1002/ccr3.5886
66. Shaw-Smith C, Willatt L, Thalange N. Growth deficiency in oculodigitoesophagoduodenal (Feingold) syndrome--case report and review of the literature. Clin Dysmorphol. 2005;14(3):155-158.
67. Aslam M, van Bokhoven H, Taylor CM. End-stage renal failure, reflux nephropathy and Feingold's syndrome. Pediatr Nephrol. 2008;23(1):159-61. doi:10.1007/s00467-007-0602-3
68. Dodds A, Ramsden R, Kingston H. Feingold syndrome--a cause of profound deafness. J Laryngol Otol. 1999;113(10):919-21. doi:10.1017/s0022215100145591
69. Kellermayer R, Weisenbach J, Gyuris P, Aszmann M, Kosztolányi G. Vertebral defects in a patient with Feingold syndrome. Clin Dysmorphol. 2005;14(4):213-214.
70. Grote LE, Repnikova EA, Amudhavalli SM. Expanding the phenotype of feingold syndrome-2. Am J Med Genet A. 2015;167A(12):3219-25. doi:10.1002/ajmg.a.37368
71. Ganjavi H, Siu VM, Speevak M, MacDonald PA. A fourth case of Feingold syndrome type 2: psychiatric presentation and management. BMJ Case Rep. 2014;2014:. doi:10.1136/bcr-2014-207501
72. Isobe A, Maeda N, Fujita H, Banno S, Kageyama T, Hatabu N, et al. Metacarpophalangeal pattern profile analysis for a 3-month-old infant with Feingold syndrome 2. Am J Med Genet A. 2021;185(3):952-954. doi:10.1002/ajmg.a.62038
73. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Polyhydramnios, p. 774.
74. Layman-Pleet L, Jackson CC, Chou S, Boycott KM. Feingold syndome: a rare but important cause of syndromic tracheoesophageal fistula. J Pediatr Surg. 2007;42(9):E1-3. doi:10.1016/j.jpedsurg.2007.06.005
75. Shaw-Smith C. Oesophageal atresia, tracheo-oesophageal fistula, and the VACTERL association: review of genetics and epidemiology. J Med Genet. 2006;43(7):545-54. doi:10.1136/jmg.2005.038158
76. Dingemann C, Eaton S, Aksnes G, Bagolan P, Cross KM, De Coppi P, et al. ERNICA Consensus Conference on the Management of Patients with Esophageal Atresia and Tracheoesophageal Fistula: Diagnostics, Preoperative, Operative, and Postoperative Management. Eur J Pediatr Surg. 2020;30(4):326-336. doi:10.1055/s-0039-1693116
77. Krishnan U, Mousa H, Dall'Oglio L, Homaira N, Rosen R, Faure C, et al. ESPGHAN-NASPGHAN Guidelines for the Evaluation and Treatment of Gastrointestinal and Nutritional Complications in Children With Esophageal Atresia-Tracheoesophageal Fistula. J Pediatr Gastroenterol Nutr. 2016;63(5):550-570. doi:10.1097/MPG.0000000000001401
78. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-24. doi:10.1038/gim.2015.30
79. Malinowski J, Miller DT, Demmer L, Gannon J, Pereira EM, Schroeder MC, et al. Systematic evidence-based review: outcomes from exome and genome sequencing for pediatric patients with congenital anomalies or intellectual disability. Genet Med. 2020;22(6):986-1004. doi:10.1038/s41436-020-0771-z
80. Huang SJ, Amendola LM, Sternen DL. Variation among DNA banking consent forms: points for clinicians to bank on. J Community Genet. 2022;13(4):389-397. doi:10.1007/s12687-022-00601-3
81. Haeusler MC, Berghold A, Stoll C, Barisic I, Clementi M. Prenatal ultrasonographic detection of gastrointestinal obstruction: results from 18 European congenital anomaly registries. Prenat Diagn. 2002;22(7):616-23. doi:10.1002/pd.341
82. Brosens E, Eussen H, van Bever Y, van der Helm RM, Ijsselstijn H, Zaveri HP, et al. VACTERL Association Etiology: The Impact of de novo and Rare Copy Number Variations. Mol Syndromol. 2013;4(1-2):20-6. doi:10.1159/000345577
83. Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 201.
84. Zhao X, Sun M, Zhao J, Leyva JA, Zhu H, Yang W, et al. Mutations in HOXD13 underlie syndactyly type V and a novel brachydactyly-syndactyly syndrome. Am J Hum Genet. 2007;80(2):361-71. doi:10.1086/511387
85. 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. 807.
86. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 194.
87. Ravel A, Chouery E, Stora S, Jalkh N, Villard L, Temtamy S, et al. How many entities exist for the spectrum of disorders associated with brachydactyly, syndactyly, short stature, microcephaly, and intellectual disability?. Am J Med Genet A. 2011;155A(4):880-4. doi:10.1002/ajmg.a.33879
88. Isidor B, Le Meur G, Conti C, Caldagues E, Lainey E, Launay E, et al. Exudative retinopathy, cerebral calcifications, duodenal atresia, preaxial polydactyly, micropenis, microcephaly and short stature: a new syndrome?. Am J Med Genet A. 2013;161A(8):1829-32. doi:10.1002/ajmg.a.36021
89. 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. 453.
90. 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. 416.
91. 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. 431.
92. 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. 782.
93. Orphanet. Orofaciodigital syndrome type 18 (ORPHA:508501). Available from: https://www.orpha.net/en/disease/detail/508501
94. Dingemann C, Eaton S, Aksnes G, Bagolan P, Cross KM, De Coppi P, et al. ERNICA Consensus Conference on the Management of Patients with Esophageal Atresia and Tracheoesophageal Fistula: Follow-up and Framework. Eur J Pediatr Surg. 2020;30(6):475-482. doi:10.1055/s-0039-3400284
95. Durkin N, Pellegrini M, Gorter R, Slater G, Cross KMK, Ure B, et al. Follow-up and transition practices in esophageal atresia: a review of European Reference Network on rare Inherited and Congenital Anomalies (ERNICA) centres and affiliates. Pediatr Surg Int. 2024;40(1):300. doi:10.1007/s00383-024-05865-z
96. Dingemann C, Eaton S, Aksnes G, Bagolan P, Cross KM, De Coppi P, et al. ERNICA Consensus Conference on the Management of Patients with Long-Gap Esophageal Atresia: Perioperative, Surgical, and Long-Term Management. Eur J Pediatr Surg. 2021;31(3):214-225. doi:10.1055/s-0040-1713932
97. Azurity Pharmaceuticals. PRILOSEC (omeprazole magnesium) for delayed-release oral suspension — Prescribing Information. NDA 022056. DailyMed set ID b6761f84-53ac-4745-a8c8-1e5427d7e179; revised 2024-03-19. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=b6761f84-53ac-4745-a8c8-1e5427d7e179
98. Grimberg A, DiVall SA, Polychronakos C, Allen DB, Cohen LE, Quintos JB, et al. Guidelines for Growth Hormone and Insulin-Like Growth Factor-I Treatment in Children and Adolescents: Growth Hormone Deficiency, Idiopathic Short Stature, and Primary Insulin-Like Growth Factor-I Deficiency. Horm Res Paediatr. 2016;86(6):361-397. doi:10.1159/000452150
99. Pfizer (Pharmacia & Upjohn). GENOTROPIN (somatropin) for injection — Prescribing Information. BLA 020280. DailyMed set ID ffebf88b-d257-4542-9808-74d9b7167765; effective revision 2026-08-04. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=ffebf88b-d257-4542-9808-74d9b7167765
100. European Medicines Agency. Omnitrope (somatropin): EPAR — Product Information. First authorisation 12 April 2006. Available from: https://www.ema.europa.eu/en/documents/product-information/omnitrope-epar-product-information_en.pdf
101. Otsuka America Pharmaceutical. ABILIFY (aripiprazole) — Prescribing Information. NDA 021436. DailyMed set ID c040bd1d-45b7-49f2-93ea-aed7220b30ac; revised 2025-01-29. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=c040bd1d-45b7-49f2-93ea-aed7220b30ac
102. European Medicines Agency. Abilify (aripiprazole): EPAR — Product Information. First authorisation 04 June 2004. Available from: https://www.ema.europa.eu/en/documents/product-information/abilify-epar-product-information_en.pdf
103. Viatris. NEURONTIN (gabapentin) — Prescribing Information. NDA 020235. DailyMed set ID 97935fd9-1d4a-43b6-a5d9-de994591187b; revised 2026-05-12. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=97935fd9-1d4a-43b6-a5d9-de994591187b
104. ClinicalTrials.gov (API v2). Searches "Feingold syndrome", "MIR17HG", "oculodigitoesophagoduodenal" and "megalencephaly polydactyly MYCN": 0 studies. Available from: https://clinicaltrials.gov/ (accessed 26 Oct 2026)
105. Segal N. Twins With Feingold Syndrome: Overview and Interview/Twin Research Summaries: Pseudoamonoamniotic Twins; Selective Fetal Growth Restriction in Monoamniotic Twins; Twin Authorship Attribution; Twins Discordant for Clubfoot/Human Interest: Twins with Different Fathers; Biological Relatedness in a Married Couple; Birth of Identical Quadruplets; Twins Switching Places. Twin Res Hum Genet. 2026;29(3):199-203. doi:10.1017/thg.2026.10080
106. Kamihara J, Bourdeaut F, Foulkes WD, Molenaar JJ, Mossé YP, Nakagawara A, et al. Retinoblastoma and Neuroblastoma Predisposition and Surveillance. Clin Cancer Res. 2017;23(13):e98-e106. doi:10.1158/1078-0432.CCR-17-0652
107. Atik T, Güvenç MS, Onay H, Özkinay F, Çoğulu Ö. GENETIC COUNSELLING IN FEINGOLD SYNDROME AND A NOVEL MUTATION. Genet Couns. 2016;27(3):381-384.
108. Rahbari R, Wuster A, Lindsay SJ, Hardwick RJ, Alexandrov LB, Turki SA, et al. Timing, rates and spectra of human germline mutation. Nat Genet. 2016;48(2):126-133. doi:10.1038/ng.3469
109. EAT — Federation of Esophageal Atresia and Tracheo-Esophageal Fistula support groups. Available from: https://www.we-are-eat.org/
110. TOFS — Tracheo-Oesophageal Fistula Support (UK). Available from: https://tofs.org.uk/
111. Unique — Rare Chromosome Disorder Support Group. Available from: https://rarechromo.org/
112. Brodeur GM, Diller LR, Nichols KE, Plon SE, Porter CC, Malkin D. Pediatric Cancer Predisposition and Surveillance Update: Summary Perspective and Future Directions. Clin Cancer Res. 2025;31(13):2581-2588. doi:10.1158/1078-0432.CCR-25-0423