Beckwith-Wiedemann Spectrum

Literature and regulatory information updated: 6 October 2026.

Beckwith-Wiedemann syndrome (BWS), Beckwith-Wiedemann spectrum (BWSp), EMG syndrome (exomphalos–macroglossia–gigantism), Wiedemann-Beckwith syndrome and isolated lateralised overgrowth (formerly isolated hemihyperplasia or hemihypertrophy) with an 11p15.5 alteration

0. Initial Reference Table

1. Introduction

Beckwith-Wiedemann syndrome is the most frequent overgrowth disorder of epigenetic origin and the clinical paradigm of genomic imprinting disorders: it results from genetic or epigenetic alterations of a gene cluster at 11p15.5 whose expression depends on the parent who transmitted each allele, and manifests as increased prenatal and postnatal growth, macroglossia, anterior abdominal wall defects, asymmetrical growth of body segments, hypoglycaemia due to hyperinsulinism and an increased risk of embryonal tumours in early childhood [1, 2]. Imprinting denotes the process whereby an epigenetic mark, mainly DNA methylation established during gametogenesis, silences one parental allele of certain genes, so that only the maternal or only the paternal allele is expressed; these marks are erased and rewritten in the germline in each generation [17].

The clinical description is attributable to two independent authors. Hans-Rudolf Wiedemann published in 1964, in the Journal de Génétique Humaine, a "familial malformation complex with umbilical hernia and macroglossia", questioning whether it was a "new syndrome" [18, 19]. J. Bruce Beckwith described in 1969 the association of macroglossia, omphalocele, adrenal cortical cytomegaly, gigantism and hyperplastic visceromegaly, based on autopsy findings; fetal adrenal cytomegaly continues to be considered pathognomonic [19]. The initial designation EMG syndrome (exomphalos, macroglossia, gigantism) reflects the original triad [1, 5]. In 1970, Thorburn and colleagues described the syndrome in Jamaican infants, in a series that still underpins the historical estimate of 1 case per 13,700 births [1, 7].

Molecular understanding advanced from the 1990s onwards. In 1993, Weksberg and colleagues demonstrated loss of IGF2 imprinting (biallelic expression of insulin-like growth factor 2) in patients with BWS, and in the same year characterisation of cytogenetic alterations of the region suggested that the causative gene was subject to imprinting [20, 21]. In subsequent years, the two 11p15.5 imprinting centres, mosaic segmental paternal uniparental disomy, CDKN1C variants and loss of methylation of imprinting centre 2 as the most frequent mechanism were identified [1, 2].

For decades there was no agreed clinical definition: proposed criteria variably combined "major" and "minor" signs, and the Oxford Desk Reference still presented suggested criteria in 2017, noting that no consensus had been reached [5]. In 2018, the European Network for Congenital Imprinting Disorders (EUCID.net), with COST funding, published the first international consensus, developed by 41 participants from 36 institutions in 11 countries, predominantly European, with 72 recommendations voted on by 33 experts through a modified Delphi process, and introduced the concept of Beckwith-Wiedemann spectrum (BWSp), encompassing classical BWS, atypical presentations with an 11p15 alteration and isolated lateralised overgrowth with an 11p15 alteration [2]. The same consensus proposed tumour surveillance stratified by molecular mechanism, which diverged from the North American recommendations of the American Association for Cancer Research (AACR), based on a 1% risk threshold and a uniform protocol; this divergence persists and was reviewed in 2024 by the AACR, in 2025 by the American College of Medical Genetics and Genomics (ACMG) for isolated lateralised overgrowth and in 2025 by the Italian BWS Association, which brought European practice closer to North American practice on some points [10–13].

The relevance of BWS to the geneticist derives from four practical reasons. Molecular diagnosis defines tumour risk and the type of tumour to seek, familial recurrence risk (below 1% in most families, but up to 50% with hereditary mechanisms) and the likely severity of hyperinsulinism [1, 22]. Somatic mosaicism, present in most sporadic cases, makes laboratory diagnosis dependent on the tissue analysed and means that a normal blood result cannot exclude the disease [2, 12]. The association with assisted reproductive technologies (ART) and monozygotic twinning raises questions about the establishment of epigenetic marks in the early embryo [2, 23]. Finally, manifestations tend to diminish with age, so diagnosis may be missed in an adult parent, with direct consequences for counselling [5].

2. Molecular Genetics and Pathophysiology

2.1 Architecture of the critical 11p15.5 region

The critical BWS region occupies approximately 1 megabase at 11p15.5–p15.4 and contains growth-regulating genes, many subject to imprinting, organised into two functionally independent domains, each controlled by its own imprinting centre (IC), corresponding to a differentially methylated region (DMR) between the maternal and paternal chromosomes [2, 5]. The 2018 consensus recommends Human Genome Variation Society (HGVS) nomenclature for these regions, H19/IGF2:IG-DMR for centre 1 and KCNQ1OT1:TSS-DMR for centre 2, although the abbreviations IC1 and IC2 remain in clinical practice; the synonyms ICR1, DMR1 or H19-DMR and ICR2, DMR2, KvDMR1 or LIT1-DMR also appear in the literature [1, 2].

Telomeric domain 1 contains IGF2, which encodes insulin-like growth factor 2, the principal fetal growth factor, expressed only from the paternal allele, and H19, which produces a maternally expressed long non-coding RNA with a probable tumour suppressor function [1, 4]. IC1 lies upstream of the H19 promoter and is methylated on the paternal allele and unmethylated on the maternal allele. The zinc-finger insulator protein CTCF binds to the unmethylated maternal allele, preventing downstream enhancers from activating IGF2 promoters; CTCF does not bind to the methylated paternal allele, IGF2 is expressed and H19 is silenced [1]. Expression of IGF2 and H19 is therefore reciprocal [1].

Centromeric domain 2 contains CDKN1C, KCNQ1 and KCNQ1OT1 [1]. IC2 is located in intron 10 of KCNQ1 and corresponds to the promoter of KCNQ1OT1, a long non-coding RNA transcribed in the antisense direction only from the paternal allele; IC2 is normally methylated on the maternal allele, silencing KCNQ1OT1 on that chromosome and allowing maternal expression of the domain's genes, including CDKN1C [1, 2]. CDKN1C encodes protein p57KIP2, an inhibitor of G1-phase cyclin-dependent kinases, a negative regulator of cell proliferation and fetal growth, with preferential maternal expression [1, 5]. KCNQ1 encodes a potassium channel subunit implicated in Romano-Ward and Jervell and Lange-Nielsen syndromes; it has maternal expression in most tissues, except the heart [1].

2.2 Molecular mechanisms and frequencies

Any alteration increasing the effective dosage of paternally expressed growth-promoting genes (IGF2) or decreasing that of maternally expressed growth-restricting genes (CDKN1C, H19) can produce the phenotype [5, 19]. The 2018 consensus summarises molecular categories, their frequency, the presence of mosaicism, recurrence risk and clinical associations in the following table, supplemented by GeneReviews and Oxford Desk Reference data [1, 2, 24].

In the population-based Piedmont cohort (Italy), among 41 tested patients, 70.7% had an identified alteration: IC2 LOM in 29.3%, upd(11)pat in 24.4%, IC1 GOM in 14.6% and a CDKN1C variant in 2.4%, with 29.3% negative results; these figures show that the distribution depends on ascertainment and methods [6]. GeneReviews notes that frequencies vary between populations [1].

2.3 Loss of IC2 methylation

IC2 LOM on the maternal chromosome leads to biallelic KCNQ1OT1 expression and reduced CDKN1C expression; it is the most frequent mechanism and results, in most cases, from a sporadic mosaic postzygotic epimutation [1, 2]. IC2 deletions are rare and the consensus does not recommend searching for single-nucleotide variants (SNVs) in these patients [2]. Approximately one-third of patients with IC2 LOM have altered methylation at other imprinted loci (MLID, multi-locus imprinting disturbance), particularly loci methylated in the maternal germline; GeneReviews gives 30–50% and a female predominance of 4:1, and the Italian Association reports 12.5–50% depending on the series [1, 2, 13]. In a Spanish cohort of 101 patients published in 2026, multilocus MS-MLPA (ME034 kit) detected MLID in 15.8%, rising to 58.4% with methylation microarrays; all cases with MLID had IC2 hypomethylation and 14.6% of MLID cases with conception data were born after ART [25]. The difference between the two methods illustrates that MLID frequency depends mainly on technical coverage.

Some MLID cases result from biallelic (sometimes heterozygous) variants in maternal-effect genes encoding proteins of the oocyte subcortical maternal complex, such as NLRP2, NLRP5, NLRP7, PADI6 and KHDC3L; these variants are present in the mother, cause pregnancy losses, hydatidiform moles and pre-eclampsia, and confer a high recurrence risk of imprinting disorders [1, 2, 13]. The 2024 international joint statement on MLID formulated 16 propositions and 8 provisional recommendations and proposed the term "multilocus imprinting disturbance syndrome" for patients with MLID and an atypical phenotype [26].

2.4 Gain of IC1 methylation

IC1 GOM on the maternal allele silences H19 and activates maternal IGF2, with biallelic IGF2 expression [1, 4]. Up to 20% of these patients have small CNVs in the DMR, undetectable by microarray, or SNVs at OCT4 or SOX binding sites, carrying a recurrence risk of up to 50% with maternal transmission, with incomplete penetrance and possible anticipation described in some families [2]. Some of these CNVs are detected by MS-MLPA; SNVs require targeted IC1 sequencing, available in specialised laboratories and prioritised when there is a family history [2].

2.5 Paternal uniparental disomy of 11p15

BWS upd(11)pat is a mosaic segmental isodisomy resulting from postzygotic somatic recombination, with increased expression of paternal growth-promoting genes (IGF2) and absence of maternal growth-restricting genes (H19, CDKN1C); it simultaneously alters IC1 (gain) and IC2 (loss) methylation [1, 2, 5]. As the proportion of disomic cells varies between tissues, the alteration may not be detected if mosaicism in the tissue analysed is low [1]. A severe phenotype usually reflects high levels of upd(11)pat mosaicism [1, 19]. In a Dutch cohort of 46 patients with upd(11)pat, clinical score correlated with the proportion of mosaicism in blood, mainly through macroglossia, but not with the length of the disomic segment [27].

In up to 10% of patients with upd(11)pat, disomy encompasses the entire genome (mosaic paternal unidiploidy or genome-wide paternal uniparental isodisomy), with signs of several imprinting disorders, more frequent developmental delay, a more severe presentation and higher tumour risk, including hepatic and adrenal tumours into adolescence; chimeric forms may have mixed cell lineages in the gonads, with a risk of germ cell tumours [1, 2]. Earlier studies described this form in 2–19% of patients with upd(11)pat, but no case was found in the 2026 Dutch series, indicating dependence on ascertainment [27]. The AACR recommends assessing other imprinting disorders, including Angelman syndrome, in children with genome-wide paternal UPD [11].

The genes ABCC8 and KCNJ11, encoding subunits SUR1 and Kir6.2 of the β-cell ATP-sensitive potassium channel (KATP), are located at 11p15.1 [28, 29]. When the paternal isodisomic segment includes these genes and the father transmitted a recessive variant, disomic cells become homozygous for the variant, producing severe hypoglycaemia; this is the same mechanism as the focal form of congenital hyperinsulinism, in which somatic loss of the maternal 11p15 allele in a β-cell clone disrupts the balance of IGF2 and H19 expression [2, 29].

2.6 CDKN1C variants

Loss-of-function variants of the maternal CDKN1C allele account for approximately 5% of sporadic cases and approximately 40% of familial cases and are associated with omphalocele, cleft palate, hypospadias, posterior fossa malformations and a positive family history [1, 2, 19]. As the gene is expressed from the maternal allele, a paternally inherited variant is usually silent; a female carrier who inherited it from her father may be clinically normal and transmit the disease to her children, as illustrated by the case described in section 10 [1, 30]. Rare cases of paternal transmission with a phenotype and maternal germline mosaicism with recurrence in siblings of apparently de novo cases have been described [1, 2].

Protein p57KIP2 has 316 amino acids in the isoform encoded by NM_000076.2 and three domains: a cyclin-dependent kinase inhibitor domain (amino acids 1–110), proline–alanine repeats (156–213) and a C-terminal QT domain that binds PCNA [30]. The MANE Select transcript of CDKN1C is NM_001122630.2; variants described in NM_000076.2 have different numbering, so conversion should be confirmed using a normalisation tool rather than calculated manually [31]. A CDKN1C variant of uncertain significance neither establishes nor excludes the diagnosis [1].

Gain-of-function CDKN1C variants, with the opposite effect, cause IMAGe syndrome (intrauterine growth restriction, metaphyseal dysplasia, congenital adrenal hypoplasia and genital anomalies), through maternal variants in the PCNA-binding domain, and have been described in a family with Silver-Russell syndrome [1, 4].

2.7 Copy number variants and rearrangements

Paternally transmitted 11p15 duplications cause BWS and maternally transmitted duplications may cause Silver-Russell syndrome; paternal duplication of the telomeric domain and maternal deletion of the centromeric domain usually cause BWS with a high recurrence risk [2]. Carriers of the balanced form of a translocation or inversion that produces a paternal duplication through unbalanced segregation have a normal phenotype [2]. Predicting the phenotype of small CNVs is complex and depends on size and the regulatory elements included [2]. Balanced maternal 11p15 translocations and inversions, with or without altered IC2 methylation, should be sought when first-line testing is negative [2]. Duplications visible on karyotyping are associated with developmental delay, and disruption of KCNQ1 by a t(11;17) translocation or an intragenic deletion was associated with congenital long QT syndrome in two families [1, 2].

2.8 Tissue pathophysiology

The phenotype results from the imbalance between IGF2 overexpression and CDKN1C loss of function, with increased proliferation in embryonic tissues; mosaicism explains why the distribution of altered cells determines the laterality of growth, internal organ involvement and the discrepancy between external phenotype and tumour risk [1, 2]. Approximately half of patients show biallelic IGF2 expression in tissues, usually with conserved maternal monoallelic H19 expression [1]. Nephrogenic rests, precursors of Wilms tumour, are typically perilobar in overgrowth syndromes, whereas intralobar rests predominate in WT1-related syndromes [32]. Multifocal or diffuse nephroblastomatosis explains the greater frequency of bilateral, synchronous or metachronous tumours in BWSp [2].

Hypoglycaemia results from excess insulin due to islet hyperplasia and, in patients with extensive upd(11)pat, also from somatic homozygosity for ABCC8 or KCNJ11 variants [5, 22]. Hyperinsulinism severity differs between subgroups, suggesting that an isolated alteration of either domain is sufficient to cause it, although to a lesser degree than upd(11)pat [22]. The placenta contributes to the phenotype: placentomegaly with weight that may reach twice normal, placental mesenchymal dysplasia and chorioangiomas [1, 5, 19].

2.9 Allelic disorders and related conditions

Molecular 11p15 alterations also occur in children with apparently isolated lateralised overgrowth (IC2 LOM, IC1 GOM, upd(11)pat), which the consensus includes within BWSp when the alteration is demonstrated, and in isolated Wilms tumours with constitutional 11p15 alterations [1, 2]. Loss of methylation of paternal IC1, the mirror of the BWS mechanism, causes Silver-Russell syndrome and isolated hemihypoplasia [1]. Mosaic and chimeric genome-wide paternal isodisomy constitute a differential diagnosis with a more severe phenotype [1]. The focal form of congenital hyperinsulinism shares with upd(11)pat the mechanism of loss of the maternal 11p15 allele restricted to the pancreas [29].

2.10 Nomenclature

GeneReviews proposes dyadic nomenclature combining mechanism and phenotype (for example, 11p15.5 IC1-related BWS, 11p15.5 UPD-related BWS or 11p15.5 IC2-related BWS), and the ACMG adopts the same approach for lateralised overgrowth (11p15.5 methylation BWS, 11p15.5 deletion BWS, CDKN1C-related BWS) [1, 12]. This model avoids grouping different tumour and recurrence risks under the same clinical name.

3. Clinical Spectrum and Manifestations by System

3.1 Overview

BWS is a spectrum: a patient may have most characteristic signs or only one or two, and mosaicism explains why expression varies between tissues and between monozygotic twins [1, 2]. Published frequencies come from retrospective series with ascertainment bias and different definitions, particularly for macrosomia, and should therefore be interpreted as approximate magnitudes [1]. The following table brings together GeneReviews values, adapted from the 2018 consensus and Wang and colleagues, with other data from the core sources.

3.2 Prenatal and perinatal period

Overgrowth begins in the second half of gestation, and pregnancy is associated with polyhydramnios, prematurity (approximately 50% according to the Oxford Desk Reference), a long umbilical cord and placentomegaly, with placentas reaching almost twice the expected weight for gestational age [1, 5, 19]. In the population-based EUROCAT study, comprising 371 cases from 34 registries in 16 European countries between 1990 and 2015, prematurity affected 40.6% of boys and 33.9% of girls, 39.9% of cases were suspected prenatally, 36.3% at birth and 16.2% in the first year, the prenatal detection rate was 64.1% (141/220), without temporal variation, and mean gestational age at ultrasound diagnosis was 19.8 ± 6.2 weeks [8]. Omphalocele is the most frequent and earliest prenatal finding; macrosomia, macroglossia, visceromegaly and retrognathia appear later, and macrosomia is rarely evident before the end of the second trimester [1, 5, 19]. Placental mesenchymal dysplasia, adrenal cysts and masses, urinary tract abnormalities and cardiac defects are rarer ultrasound findings; increased free human chorionic gonadotrophin in the first trimester and maternal serum α-fetoprotein (αFP) in the second trimester, the latter associated with omphalocele, may accompany fetal BWSp [2, 5].

Maternal complications include gestational hypertension, with approximately 2.4-fold higher risk, pre-eclampsia and, occasionally, HELLP syndrome (haemolysis, elevated liver enzymes and thrombocytopenia) [2]. Macrosomia confers a risk of shoulder dystocia [2]. Infant mortality, historically estimated at approximately 20–21%, results from prematurity, airway obstruction due to macroglossia, hypoglycaemia and, rarely, cardiomyopathy; GeneReviews considers this figure an overestimate in light of current recognition and treatment [1, 19]. In EUROCAT, 93.3% of pregnancies ended in live birth, 4.0% in termination of pregnancy and 2.7% in fetal death; 3.6% of the 330 liveborn infants with information died in the first week (11 of the 12 were premature), and first-year survival was 90.9% [8]. Neonatal adaptation may include apnoea, cyanosis and seizures, and neonatal polycythaemia is occasional and self-limiting [1, 19].

3.3 Overall growth

Birth weight is approximately 4 kg (mean length 52.6 cm) and is usually near the 97th centile, although overgrowth is not constant [5, 19]. Growth rate remains high in early childhood and slows around 7–8 years; adult height usually lies in the upper half of the normal range [1, 5]. Growth should be interpreted relative to that of the parents: a height at the 85th centile may represent overgrowth if the parents are near the 10th centile [1]. Final height data partially challenge the idea of normal adult height: the consensus cites a study with final height 1.7 ± 1.1 standard deviations above parental target height and approximately half of patients above +2 standard deviations, and syndrome-specific charts constructed from 581 patients from the Netherlands, Italy and the USA estimated mean height at 18 years at 180.6 cm in boys (+0.6 standard deviations) and 166.3 cm in girls (+0.5), with body mass index above +1 standard deviation at all ages, slightly greater height in the IC1 subgroup and lower height in upd(11)pat [2, 13]. Accelerated bone maturation is more marked in the first 4 years and rarely persists; the consensus estimates advanced bone age in approximately 3% [2, 19]. Puberty occurs at the usual age and macrocephaly is not typical [1, 19].

3.4 Lateralised overgrowth

Lateralised overgrowth, a term replacing hemihypertrophy and hemihyperplasia, is increased length and/or circumference of parts of one side of the body relative to the contralateral side; it may cross the midline, involve ipsilateral or contralateral segments and include internal organs [1, 12]. In BWS, muscle enlargement predominates, sometimes with bone growth [1]. Limb length discrepancies up to 1 cm may be physiological, and a difference greater than 1 cm accompanied by a measurable circumference difference should be considered abnormal [1, 33]. The ACMG describes as suggestive of BWSp a pattern of intrasegmental proportionality with intersegmental disproportion (for example, a foot and lower leg proportionate to each other but larger than the thigh and contralateral limb), present from birth or noticed by 6–12 months, stable during childhood and less apparent in adolescence; intrasegmental disproportion (digits of different sizes) points to the PIK3CA spectrum and distorting progression to Proteus syndrome [12].

Lateralised overgrowth occurs in all subgroups, is the most frequent manifestation in upd(11)pat and is rare with CDKN1C variants [2]. In a series of 142 children with a molecular diagnosis, limb volume asymmetry progressed slowly in some regions in the IC2 subgroup (0.47 mm/year in the arm, 0.53 mm/year in the lower leg and 0.40 mm/year in foot length), upd(11)pat had greater baseline asymmetry and progression rates were similar between genotypes [34]. The Italian Association classifies lower limb length discrepancy as minimal (<1 cm), mild (1–2 cm), severe (2–5 cm) and critical (>5 cm) and reports that half of discrepancies below 1 cm progress to mild (40%) or severe (12.5%), 35% of mild discrepancies to severe and 15.4% of severe discrepancies to critical, with less favourable progression in upd(11)pat [13].

3.5 Craniofacial region

Macroglossia (approximately 90%) represents a true increase in tongue volume in length, width and thickness and should be distinguished from protrusion due to hypotonia [1, 2]. It may obstruct the neonatal airway, impair feeding, cause drooling, speech articulation abnormalities, anterior open bite and prognathism [1, 2]. Many patients do not require surgery because growth of the oral cavity accommodates the tongue; approximately 40% undergo tongue reduction [1, 2]. Facial features include infraorbital creases, midface retrusion, a thin upper lip vermilion and prominent mandible, prominent eyes due to infraorbital hypoplasia, nevus simplex of the forehead, glabella and eyelids, a metopic ridge, large fontanelles and a prominent occiput; features tend to normalise over time [1, 19]. Horizontal linear earlobe creases are anterior and pits occur on the posterior helical rim; earlobe creases appearing in adulthood are not part of the condition, and posterior helical pits are rarely found outside BWS [1, 19]. Cleft palate is rare and associated with CDKN1C [1].

3.6 Abdominal wall

Omphalocele, umbilical hernia and diastasis recti are frequent; large umbilical hernias may contain bowel and require surgical correction [1]. Omphalocele is preferentially associated with centromeric domain alterations (IC2 LOM and CDKN1C) and is rare in IC1 GOM and upd(11)pat [2]. Smith's also mentions posterior diaphragmatic eventration and cryptorchidism [19].

3.7 Glucose metabolism

Hypoglycaemia in BWSp is due to excess insulin and occurs in 30–60% of children [1, 2]. Most episodes are mild and resolve within 72 hours or less; in up to 20% of newborns it persists beyond the first week and requires medical treatment or, in severe cases, pancreatectomy, and onset may be delayed, occurring during the first month [1, 2]. The consensus distinguishes hyperinsulinism (a cardinal sign when lasting more than one week and/or requiring treatment escalation) from transient hypoglycaemia (a suggestive sign when lasting less than one week) [2]. In a cohort of 501 patients with hyperinsulinism, excluding focal forms, approximately 6% had BWSp signs, most with upd(11)pat, and half of these underwent surgery for persistent hypoglycaemia despite optimised medical treatment [2]. Saudubray lists BWS, together with Kabuki and Turner syndromes, among the most frequent syndromic causes of congenital hyperinsulinism, usually responsive to diazoxide and resolving during the first year of life [29].

A North American cohort of 85 children with BWS and hyperinsulinism (2009–2024), published in 2026, showed significant differences between subgroups: median maximum glucose infusion rate was 16.2 mg/kg/min in upd(11)pat, 12.0 in IC2, 30.0 in genome-wide paternal UPD, 10.0 in IC1 and 4.2 in the two patients with CDKN1C; the need for continuous enteral dextrose was 57.9%, 22.7%, 100%, 33.3% and 50%, and pancreatectomy was performed in 46.3%, 8.3%, 33.3%, 0% and 0%, respectively [22]. All patients whose isodisomy included the ABCC8/KCNJ11 region and who had a pathogenic variant in one of these genes were unresponsive to diazoxide and underwent surgery; most patients with IC2 LOM responded to diazoxide (12 of 15), although one patient with a mosaic GCK variant in the pancreas required surgery [22]. Hypothyroidism has been described but is rare [1].

3.8 Embryonal tumours

Approximately 8% of children with BWSp develop embryonal tumours: Wilms tumour (52% of tumours), hepatoblastoma (14%), neuroblastoma (10%), rhabdomyosarcoma (5%) and adrenal carcinoma (3%); phaeochromocytoma, pancreatoblastoma, gonadoblastoma and juvenile fibroadenoma have also been described [2, 19]. Overall risk is highest in the first 2 years and gradually decreases until puberty [2]. Wilms tumour risk is concentrated in the first 7 years (median diagnostic age approximately 24 months, with cases reported after 7 years) and hepatoblastoma risk in the first 2–4 years, always before 30 months in the Maas series; the Oxford Desk Reference reports that 96% of tumours in Beckwith's series (121 patients) occurred by 8 years [1, 5, 9]. The relative risk of hepatoblastoma is more than 2000 times that in the population [2, 11].

Risk depends on mechanism. The two major 2016 syntheses obtained similar but non-identical values: in Mussa's meta-analysis (1370 patients, 7.4% with neoplasms), tumour prevalence was 2.5% in IC2, 13.8% in upd(11)pat, 22.8% in IC1 and 8.6% with CDKN1C; in the Maas series (1971 patients), it was 28% in IC1, 16% in upd(11)pat, 6.9% with CDKN1C, 2.6% in IC2 and 6.7% without an identified defect [9, 35]. Smith's presents ranges attributing the highest values to upd(11)pat; primary data attribute them to IC1, and this is the attribution followed in this article [9, 19, 35].

Table sources: 2018 consensus (Table 3, based on Maas and Mussa), GeneReviews (Table 3, 2023 revision), George and colleagues' review for CDKN1C (113 literature cases and registry cohort) and AACR 2024 [1, 2, 11, 36]. AACR 2024 summarises Wilms tumour risk as 20% in IC1, 8% in upd(11)pat, approximately 1% in IC2, well below 1% with CDKN1C and 4–10% in the others, and overall BWSp risk as 8–12%, with 47% Wilms tumours and 25% hepatoblastomas [11]. Three limitations affect these figures: they derive from retrospective data with ascertainment bias, MS-MLPA may misclassify subgroups, and the blood epigenotype does not always match that of the tumour or its organ of origin [1, 11]. Risk is not proportional to the number of signs: mild forms and cases with negative blood testing also develop tumours, and a Wilms tumour may be the first manifestation of unrecognised BWS [1, 13]. Genome-wide paternal UPD probably carries a higher risk than upd(11)pat, with later tumours [2, 11].

In isolated lateralised overgrowth, Hoyme and colleagues' prospective study (168 children) observed 10 tumours in 9 patients (5.9%), including Wilms tumour in 3%; a retrospective review of 250 patients over 10 years (Dempsey) found 1.2% (two Wilms tumours and one adrenal carcinoma), and combining the two cohorts gives 3.1% (13/418) [12, 33, 37]. When lateralised overgrowth is associated with upd(11)pat, risk has been estimated at 32–50% [2]. The ACMG concludes that, after a complete negative molecular investigation and a score below 4, lifetime risk is probably below 5% [12].

3.9 Kidney and urinary tract

Nephrourological abnormalities affect 28–61% of patients: nephromegaly, medullary dysplasia, medullary sponge kidney, collecting system duplication, cortical and medullary cysts (approximately 10%), diverticula, nephrocalcinosis and stones, megaureter, vesicoureteral reflux, posterior urethral valves and megacystis [1, 2, 19]. In a series of 18 patients, 22% had hypercalciuria and 11% nephrocalcinosis [1]. Nephromegaly may mark an increased Wilms tumour risk, and severe vesicoureteral reflux may cause renal damage and recurrent infections [2, 19].

3.10 Heart and vessels

Cardiomegaly was described in approximately 20% before the molecular era and usually resolves without treatment [1, 5]. Congenital heart defects affect 13–20% (patent ductus arteriosus, patent foramen ovale, atrial and ventricular septal defects, among others); cardiomyopathy is rare but may be severe, and the Oxford Desk Reference estimates that BWS accounts for approximately 1% of hypertrophic cardiomyopathies diagnosed before 10 years [2, 5, 38]. Long QT syndrome is associated with rearrangements disrupting KCNQ1 [1, 2]. Smith's reports a risk of aneurysmal arterial dilatation in adults, and the Italian Association relates these to CDKN1C variants [13, 19].

3.11 Neurodevelopment and behaviour

Cognitive development is usually normal; delay is associated with prematurity, severe untreated hypoglycaemia, hypoxia, unbalanced chromosomal rearrangements (particularly 11p15 duplications) and genome-wide paternal UPD [1, 2, 5]. The longitudinal study by Weng and colleagues (15 children) showed mental and social development comparable to that of siblings and cousins [5]. Posterior fossa malformations (including Dandy-Walker) and corpus callosum or septum pellucidum malformations are rare and associated mainly with IC2 alterations and CDKN1C [1, 2]. Reported findings include language and gross motor difficulties in preschool children, slower reading and mental arithmetic at school age, parent-reported autism spectrum disorder, learning difficulties reported by approximately 20% of adults and symptoms of anxiety, low self-esteem and social isolation that appear to increase with age [1, 13]. These observations derive from small samples and questionnaires, without robust comparison groups.

3.12 Other manifestations

Hearing loss is rare, sensorineural or conductive due to stapes fixation [1]. Cutaneous and visceral haemangiomas, particularly hepatic, are more frequent than in the general population and are included in the differential diagnosis of hepatic lesions detected during surveillance [1, 10]. Smith's also describes enlarged ovaries and testes, bicornuate uterus, hypospadias, clitoromegaly, immunodeficiency, cardiac hamartoma and extravillous trophoblast cytomegaly [19]. Male subfertility and azoospermia have been reported in adulthood, probably related to late correction of cryptorchidism, and a 2026 report described six adolescent or adult women with benign or malignant breast tumours, without a defined risk or established screening utility [1, 39, 40].

4. Clinical Recognition and Warning Signs

Recognition frequently begins on obstetric ultrasound. Omphalocele, particularly when apparently isolated, placentomegaly or placental mesenchymal dysplasia, polyhydramnios, visceromegaly, nephromegaly with hyperechoic kidneys and late-gestation macroglossia should prompt consideration of BWSp; approximately 10–20% of fetuses with isolated omphalocele and approximately 20% of placental mesenchymal dysplasia cases have BWSp [2, 41–43]. Fetal macrosomia with a small omphalocele appears in the Oxford Desk Reference list of polyhydramnios causes [43]. When fetal BWSp is suspected, the consensus recommends anticipating maternal (pre-eclampsia) and fetal complications and planning delivery in a centre with neonatal intensive care [2].

In the newborn, the combination of macrosomia, macroglossia, an abdominal wall defect, earlobe creases or posterior helical pits and hypoglycaemia should prompt immediate blood glucose monitoring, because detecting and treating hypoglycaemia determines neurological prognosis [2, 19]. Omphalocele in any newborn warrants blood glucose surveillance because of its association with BWS [41]. Macroglossia may complicate intubation and anaesthesia, which should be anticipated before omphalocele surgery [2, 5].

At the genetics consultation, the history should include a three-generation pedigree with birth weights and heights of parents and siblings, familial overgrowth, abdominal wall defects and neonatal hypoglycaemia, with particular attention to the maternal line; recurrent pregnancy losses and hydatidiform moles in the mother; conception by ART; and pregnancy history (ultrasound, polyhydramnios, prematurity, placental weight) [1, 5, 24]. Examination assesses overall growth (weight, length, head circumference), macroglossia, ear creases and pits (examining the posterior helical rim), forehead nevus simplex, facial and limb asymmetry, measured using bony landmarks and compared with the contralateral side, and the abdominal wall [5, 24, 33]. In the parents of an affected child, ear examination and review of childhood photographs may reveal a diagnosis no longer evident in adulthood [1, 5].

Warning signs requiring immediate action are: hypoglycaemia persisting beyond 72 hours, a high glucose infusion rate or a need for treatment escalation (suspected hyperinsulinism); airway obstruction or obstructive sleep apnoea due to macroglossia; an abdominal mass, haematuria, abdominal distension or parental concerns during the years of greatest tumour risk; and warning signs of another condition, such as developmental delay without prematurity, hypoglycaemia or a cerebral abnormality, requiring reconsideration of the diagnosis [1, 2]. Cleft palate, advanced bone age, polydactyly or supernumerary nipples point towards alternative diagnoses such as Simpson-Golabi-Behmel syndrome, and these signs are therefore not included in the consensus score [2]. A child with a single cardinal sign, such as macroglossia, hyperinsulinism, multifocal Wilms tumour or a typical histopathological finding, should be referred to a BWS specialist even without reaching the diagnostic score [2]. Every child with Wilms tumour or hepatoblastoma should undergo genetic assessment, because 16–30% of these paediatric tumours show typical BWSp alterations [11, 13].

5. Diagnosis: Criteria, Tests and Algorithms

5.1 The 2018 international consensus scoring system

The consensus defined cardinal signs, each worth 2 points, and suggestive signs, each worth 1 point; the following table reproduces Table 1 and its note in full [2].

Original table note: clinical diagnosis of classical BWS requires a score ≥4, which does not require molecular confirmation of an 11p15 abnormality; patients with a score ≥2 (including those with classical BWS with ≥4) merit genetic testing for investigation and diagnosis; patients with a score <2 do not meet genetic testing criteria; patients with a score ≥2 and negative genetic testing should be considered for an alternative diagnosis and/or referred to a BWS expert [2].

The recommendations accompanying the table clarify the definitions. BWSp is usually caused by dysregulation of the imprinted 11p15 region, involves overgrowth in several tissues, frequently mosaic, and includes three subsets: classical BWS, lateralised overgrowth with an 11p15 abnormality and patients with an 11p15 abnormality who do not meet the preceding criteria (R1, A+++) [2]. Increased height and/or weight is no longer a cardinal sign; with a score ≥4, a patient has BWSp even without an identified 11p15 abnormality (R2, A+++) [2]. BWSp lateralised overgrowth is defined as a marked increase in length and/or circumference of most or all of one side of the body, with an 11p15 abnormality; lateralised overgrowth without an 11p15 abnormality in tested samples falls outside the consensus scope (R3, A+++) [2]. Molecular testing is recommended with ≥2 points, discretionary in isolated omphalocele, recommended when a parent has a hereditary 11p15 abnormality (50% risk) and begins with blood lymphocytes (R4, A+++), unless there is an alternative explanation for the sign (for example, gestational diabetes for macrosomia) [2]. Hypoglycaemia is defined as plasma glucose <50 mg/dL in the first 6 hours of life and <60 mg/dL thereafter; hyperinsulinism as a glucose infusion rate ≥8 mg/kg/min, detectable insulin and/or C-peptide and undetectable ketone bodies and free fatty acids; transient hypoglycaemia (a suggestive sign) meets these criteria for less than one week and cardinal hyperinsulinism for more than one week and/or with treatment escalation (R5, A++) [2]. Recommendation strength reflects the proportion of votes ("+" 26–49%, "++" 50–69%, "+++" ≥70%), rather than evidence quality [2].

In subsequent validation, the system showed 85% sensitivity and 43% specificity for identifying patients to test, with variation between ethnic groups, and does not predict tumour risk [13]. In 2025, the Italian Association proposed adding two 1-point suggestive signs: pregnancy achieved through ART and monozygotic twinning; GeneReviews already includes both among tier 3 findings [1, 13]. This proposal had not been incorporated into a new international consensus at the date of this article.

5.2 GeneReviews criteria

GeneReviews uses a three-tier classification and stresses that children with mild presentations, below the threshold or with non-diagnostic testing may have increased tumour risk [1]. Tier 1 findings are macroglossia, omphalocele, embryonal tumour (unilateral or bilateral Wilms tumour, hepatoblastoma or nephroblastomatosis), hemihyperplasia, macrosomia, hyperinsulinaemic hypoglycaemia, adrenal cortical cytomegaly, placental mesenchymal dysplasia or pancreatic adenomatosis and family history; tier 2 findings are visceromegaly, ear creases or pits, characteristic facies, renal abnormalities, a large umbilical hernia requiring surgery, other embryonal tumours (rhabdomyosarcoma, neuroblastoma, adrenal tumours) and transient hypoglycaemia requiring intervention; tier 3 findings are a small umbilical hernia or diastasis recti, polyhydramnios and/or placentomegaly, prematurity, nevus simplex and/or haemangioma, transient hypoglycaemia without intervention, structural cardiac abnormalities or cardiomegaly, ART or parental subfertility and monozygotic twinning [1]. Clinical diagnosis is established with two tier 1 findings or one tier 1 and one tier 2 finding; molecular diagnosis with at least one tier 1 or 2 finding and a BWS-associated epigenetic or genomic 11p15.5 alteration, an 11p15.5 CNV or a pathogenic or likely pathogenic CDKN1C variant [1]. A constitutional 11p15.5 alteration without clinical signs is insufficient for clinical diagnosis but confers tumour risk and warrants surveillance [1]. The pre-consensus Oxford Desk Reference criteria (three signs, two major and one minor) already recommended considering tumour surveillance in children with fewer than three signs [5].

5.3 Molecular algorithm

The consensus recommends testing by professionals experienced in imprinting disorders, using HGVS nomenclature in reports (R7, A+++), and following the flowchart in its Figure 3 (R8, A+++) [2]. First-line testing is simultaneous IC1 and IC2 methylation analysis; if altered, DMR copy number should be determined if the technique does not provide it, to distinguish epimutation, CNV and upd(11)pat (R9, A+++) [1, 2]. MS-MLPA (methylation-specific MLPA) is the most widely used method because it measures methylation and copy number simultaneously, but methylation-specific PCR (MS-PCR) and its quantitative form (MS-qPCR) are more sensitive for low-level mosaicism; the ACMG also mentions allele-specific methylated multiplex quantitative real-time PCR (ASMM-qRT-PCR) and digital PCR [2, 12]. Simultaneous IC1 gain and IC2 loss without a CNV suggest upd(11)pat, which can be confirmed by microsatellites or SNP microarray; SNP microarray is the most sensitive method for 1–5% mosaicism and distinguishes segmental from genome-wide UPD (R10, A+++) [2]. With IC1 GOM without a CNV, detailed DMR analysis (SNVs and small CNVs) should be considered, particularly with a family history (R11, A+++) [2].

A negative blood result does not exclude BWSp (R12–R15, A+++). The next step is chosen according to phenotype: sequencing of CDKN1C coding exons and exon–intron junctions, with deletion and duplication analysis, when there is a family history, cleft palate or an abdominal wall defect; analysis of other tissues when asymmetry predominates; searching for rare chromosomal rearrangements; and reassessment of the clinical diagnosis [1, 2]. Alternative tissues include buccal swabs, fibroblasts and mesenchymal cells obtained during surgery (tongue from glossectomy, skin from the larger limb, peritumoural tissue); a buccal swab taken from the overgrown side increases yield by 7–10% and skin biopsy by a further approximately 10% [1, 13]. The ACMG advises against cell culture for methylation analysis to avoid culture-induced changes [12]. At least 70% of patients with isolated lateralised overgrowth remain without molecular confirmation when only blood is tested [13]. In negative cases meeting criteria, the Italian Association suggests exome or genome sequencing and microarray to seek alternative or dual diagnoses (2–7.5% of syndromic patients) [13].

GeneReviews states that current techniques identify one of the alterations in more than 80% of patients with a clinical diagnosis; Smith's reports clinical testing available for approximately 80% of mechanisms [1, 19]. The sensitivity of each method refers to the proportion of known alterations detected in patients with a clinical diagnosis and should not be confused with clinical sensitivity, because approximately 20% of patients with a characteristic phenotype remain without a molecular diagnosis [1, 2]. Next-generation sequencing methods after treatment distinguishing methylated and unmethylated DNA are under investigation to detect low proportions of epimutated cells, as are bisulphite sequencing panels targeting 63 DMRs (ImprintSeq); none of these methods yet has an established place in the diagnostic algorithm [13].

5.4 Laboratory report content

The report should state the tissue analysed, method and detection limit, methylation result for each DMR using HGVS nomenclature (H19/IGF2:IG-DMR and KCNQ1OT1:TSS-DMR), copy number, interpretation of the mechanism (epimutation, CNV, segmental or genome-wide upd(11)pat) and limitations, particularly the inability to exclude low-level mosaicism [1, 2]. For CDKN1C, the report should state the gene, transcript with version, preferably MANE Select NM_001122630.2, the variant in HGVS nomenclature at coding DNA and protein levels, zygosity, classification according to ACMG/AMP criteria and, when historical numbering is used (NM_000076.2), the correspondence confirmed using a normalisation tool [1, 31]. The distinction between a pathogenic variant and one of uncertain significance has direct consequences: partial segregation in a family or consistency with the phenotype does not turn a VUS into a diagnostic variant [1].

5.5 Prenatal diagnosis

Prenatal diagnosis has two main indications: families with a known genetic alteration and high recurrence risk, and pregnancies without a family history in which ultrasound shows possible BWSp signs (R18, A+++) [2]. Genomic alterations (CDKN1C variant, CNV, rearrangement) can be sought in chorionic villi or amniotic fluid, and preimplantation genetic diagnosis is possible for familial CDKN1C variants and some genomic alterations, but not for methylation status [1]. Methylation analysis is more complex: amniotic fluid is considered the most reliable sample, chorionic villi may not reflect the fetus and have produced false positives, culture may alter methylation and false negatives occur because of mosaicism, so a normal result does not exclude the diagnosis (R19–R20, A+++) [1, 2]. The Italian Association recommends sampling after 13–14 weeks because methylation may not yet be complete earlier, prefers native to cultured villi, accepts cultured amniocytes and requires same-tissue controls for MLPA [13]. Prenatal yield is lower than postnatal yield (approximately 25%), and the spectrum differs: IC2 LOM in approximately 70% and CDKN1C in approximately 13% of prenatally detected cases, owing to the association with omphalocele [13]. In a real-world series of 646 prenatal samples from three European laboratories, the detection rate was 9.75% and the technical failure rate 4.3%, almost entirely in native samples [44].

The Italian Association proposes a prenatal score, adapted from Carli and colleagues, with testing indicated from 3 points: macroglossia, omphalocele, placental mesenchymal dysplasia, family history of hereditary BWS and typical tumours or hyperplasias are worth 3 points; macrosomia (abdominal circumference >2 standard deviations, disproportionate to the femur and head circumference), visceromegaly, polyhydramnios and placentomegaly are worth 2 points; nephroureteral abnormalities, elevated chorionic gonadotrophin, αFP or PAPP-A, monozygotic twinning and pregnancy through ART are worth 1 point [13]. With apparently isolated omphalocele without a family history, GeneReviews suggests methylation testing in amniocytes, followed by CDKN1C if negative, microarray for 11p15 CNVs and serial ultrasound [1]. The consensus recommends prospective multicentre audit of prenatal results (R21, A+++) [2].

5.6 Assessments after diagnosis

After diagnosis, GeneReviews recommends: measurement of weight, length and head circumference; pre-feed blood glucose in the newborn, with investigation for hyperinsulinism if hypoglycaemia persists beyond 72 hours; assessment of macroglossia, cleft palate and the airway, with a sleep study if obstructive apnoea is suspected; abdominal wall assessment; baseline complete abdominal ultrasound, because renal ultrasound alone does not assess the liver; serum αFP interpreted using BWS-specific ranges; clinical assessment of asymmetry with orthopaedic referral if discrepancy exceeds 1 cm; audiogram; cardiovascular assessment with electrocardiogram and echocardiogram if there is clinical suspicion or a genomic alteration involving KCNQ1; language and developmental assessment, with brain magnetic resonance imaging if delay is significant; and genetic counselling [1]. The consensus adds baseline cardiovascular examination in all patients, with echocardiography if there is suspicion (R45, A+++), and clinical and ultrasound nephrourological assessment at diagnosis (R51, A+++) [2].

Clinical suspicion (score ≥2) or suggestive prenatal finding
  └─ IC1 + IC2 methylation in blood (with copy number)
       ├─ Isolated IC2 LOM → epimutation (MLID if other signs or maternal history)
       ├─ IC1 GOM → exclude DMR CNV/SNV (particularly if familial)
       ├─ IC1 GOM + IC2 LOM without CNV → upd(11)pat → SNP microarray (segmental vs genome-wide)
       ├─ DMR CNV → microarray, karyotype/FISH, parental testing
       └─ Normal → CDKN1C (family history, cleft, wall defect)
                    → other tissues (asymmetry, macroglossia)
                    → balanced rearrangements
                    → alternative diagnoses (exome/genome, overgrowth panel)
       Score ≥4 without an identified alteration = classical BWS (continue surveillance)

This flowchart summarises consensus Figure 3 and recommendations R9–R15, with additions from GeneReviews and the Italian Association [1, 2, 13].

6. Differential Diagnosis

The differential diagnosis is organised around three clinical situations: global overgrowth in the newborn or infant, lateralised overgrowth and isolated findings (omphalocele, hyperinsulinism, embryonal tumour). The two major molecular families of overgrowth syndromes are those involving epigenetic regulators (NSD1, EZH2, DNMT3A) and the PI3K/AKT/mTOR pathway (PIK3CA, AKT1, AKT3, PIK3R2, PTEN), together with imprinting disorders [24]. When a child with a presumptive BWS diagnosis has developmental delay with normal chromosome testing, no hypoxia or hypoglycaemia and normal brain imaging, other causes should be sought [1, 2].

In lateralised overgrowth, the first distinction is between overgrowth of one side and hypoplasia of the opposite side: hemihypoplasia is more frequent in children with chromosomal mosaicism, delay and growth restriction, and asymmetry with growth restriction and relative macrocephaly points towards Silver-Russell syndrome [1, 33]. Asymmetries of the face, lower limbs or chest should also be distinguished from plagiocephaly, congenital hip dislocation and chest wall deformities [1].

With isolated findings, the reasoning differs. Isolated omphalocele is more frequent and less often associated with an 11p15 defect than other isolated signs, so the consensus does not include it in BWSp and leaves testing to the physician's discretion; it is associated with aneuploidy in some cases, and BWS diagnosis should be considered if macrosomia, macroglossia, nephromegaly, a large placenta or polyhydramnios is present [2, 41]. Isolated congenital hyperinsulinism due to ABCC8, KCNJ11, GCK, GLUD1, HADH, HNF1A or HNF4A variants and the focal form due to paternal isodisomy with a KATP variant should be distinguished; the consensus suggests considering 11p15.5 methylation analysis in hyperinsulinism without other BWSp signs and, with severe persistent hyperinsulinism in a patient with BWSp, seeking additional causes (R43, A+++) [2, 22, 29]. In Wilms tumour, the differential includes WT1-related syndromes (WAGR, Denys-Drash, Frasier), familial Wilms tumour, mosaic variegated aneuploidy (BUB1B), Fanconi anaemia D1 and N (BRCA2, PALB2), Bloom syndrome, hyperparathyroidism–jaw tumour syndrome, 2q37 deletion, trisomy 18 and Mulibrey nanism [11, 32].

7. Treatments and Risk Management Strategies

7.1 General principles

There is no therapy targeting the imprinting defect. Management consists of treating manifestations, tumour surveillance and counselling, and the consensus recommends that each patient have a responsible professional coordinating referrals and care (R22, A+++) [1, 2]. Recommendations derive mainly from expert consensus: the 2018 process graded strength by voter agreement, and AACR and ACMG guidelines explicitly state that they are based on retrospective data and expert opinion, without randomised trials [2, 11, 12]. The principal guideline divergences concern tumour surveillance and are presented in section 7.8.

7.2 Pregnancy and delivery

When BWSp is suspected or confirmed prenatally, fetal complications (congenital abnormalities, shoulder dystocia due to macrosomia, postnatal hypoglycaemia) and maternal complications (pre-eclampsia) should be anticipated, and delivery should occur in a unit with neonatal intensive care (R23–R24, A+++) [2]. The Oxford Desk Reference recommends referral to maternal–fetal medicine because of the risks of prematurity, polyhydramnios, pre-eclampsia and macrosomia [5]. Isolated omphalocele may be delivered vaginally in a hospital with a neonatal unit; large omphaloceles should be delivered in a specialist perinatal centre, sometimes by caesarean section, and the newborn should undergo blood glucose surveillance and transfer to paediatric surgery [41].

7.3 Hypoglycaemia and hyperinsulinism

The consensus recommends capillary blood glucose monitoring during the first 48 hours in all newborns with suspected or diagnosed BWSp; hypoglycaemia is defined by two consecutive values, 30 minutes apart, below 50 mg/dL (2.75 mmol/L) in the first 6 hours or below 60 mg/dL (3.5 mmol/L) thereafter, and requires transfer to neonatal intensive care (R40, A++) [2]. Before discharge, at 48 hours of life, a diagnostic fasting test should be performed with glucose, insulin and ketone body measurement after a 6-hour fast in term newborns and a 4-hour fast in preterm newborns (R41, A++) [2]. Confirmed hypoglycaemia should be treated to maintain plasma glucose above 3.9 mmol/L (70 mg/dL); GeneReviews specifies oral feeding in mild cases and 10% or 25% intravenous glucose in more severe cases, targeting 3.9–5.5 mmol/L (70–100 mg/dL), and notes that, as onset may be delayed, parents should know the symptoms of hypoglycaemia [1, 2]. Hyperinsulinism diagnosis rests on detectable insulin, suppressed β-hydroxybutyrate and free fatty acids and a glycaemic response to glucagon during hypoglycaemia, and should be made with paediatric endocrinology [2]. In Saudubray, a glucose infusion requirement above 8 mg/kg/min in the newborn (above 10 in the infant) to maintain blood glucose above 3 mmol/L is characteristic of insulin-mediated hypoglycaemia, and a blood glucose rise above 1.7 mmol/L (30 mg/dL) 30–40 minutes after 1 mg glucagon during hypoglycaemia supports the diagnosis [29].

No specific management of hyperinsulinism in BWSp has been proposed; general recommendations apply (R42, A++), and additional causes should be sought in the presence of severe persistent hyperinsulinism (R43, A+++) [2]. Options include diazoxide, somatostatin analogues (octreotide, lanreotide) and, when maximum medical therapy fails, pancreatectomy; the mTOR inhibitor sirolimus and glucagon-like peptide 1 receptor (GLP-1R) antagonists have been used in hyperinsulinism and occasionally in BWSp, without specific evaluation [1, 2]. Saudubray describes the general approach to congenital hyperinsulinism: in the newborn, urgent correction of severe hypoglycaemia with 10% intravenous or oral dextrose, 2 mL/kg every 5–10 minutes (no maximum dose defined in the source), or emergency glucagon 0.3 mg/kg intramuscularly, intravenously or subcutaneously (maximum 1 mg), and continuous infusion started at 8 mg/kg/min and titrated every 15–30 minutes until normoglycaemia (no maximum rate defined in the source), often through a central catheter; continuous subcutaneous or intravenous glucagon (1–2 mg/24 h) when glucose requirements are extreme (for example, above 16 mg/kg/min), reducing glucose supply by approximately 40%; diazoxide as first-line treatment after several days (for example, 7–15 days in the newborn) without spontaneous improvement, with a 5-day response test (absence of hypoglycaemia on a normal diet and during a 6-hour fast in the newborn and a 12-hour fast in the child); and octreotide in non-responders [29].

The 2026 cohort showed that molecular subgroup predicts response: patients with upd(11)pat whose isodisomy extends to the ABCC8/KCNJ11 region had a very low probability of diazoxide response (odds ratio 0.087 relative to IC2) and greater need for pancreatectomy (odds ratio 5.00), and all those also carrying a pathogenic KATP variant underwent surgery; patients with IC2, IC1 and CDKN1C mostly responded to diazoxide [22]. The authors propose, for any child with hyperinsulinism and at least one additional BWS sign, a multigene hyperinsulinism panel and mosaicism-adapted 11p15 methylation testing (with alternative tissues, including resected pancreatic tissue), and SNP microarray to define upd(11)pat extent [22]. In one case of genome-wide paternal UPD undergoing approximately 95% pancreatectomy at 2 months, the patient still required lanreotide and pancreatic enzymes 13 years later [22]. In diffuse hyperinsulinism, subtotal pancreatectomy (95–98%) almost always leads to insulin-dependent diabetes, affecting 91% of patients over 14 years, and should be avoided whenever possible [29].

7.4 Medicines used

The American label diazoxide dose for newborns and infants (8–15 mg/kg/day) differs slightly from Saudubray's (10–15 mg/kg/day); the difference lies in the starting dose, and the maximum of 15 mg/kg/day is the same [14, 29]. Somatostatin analogues have no approved hyperinsulinism indication in the sources consulted, and their use in BWS extrapolates from congenital hyperinsulinism practice, with dosage supported by Saudubray [15, 29].

7.5 Macroglossia and airway

With suspected significant airway obstruction, assessment by polysomnography and respiratory medicine and otorhinolaryngology consultation should be undertaken (R33, A+++); with severe obstruction, adenotonsillectomy and/or tongue reduction should be considered earlier (R35, A+++), and, rarely, neonatal surgery requires prior tracheostomy [2]. Without obstruction, tongue reduction is usually considered after the first year of life, when tongue size stabilises, for feeding difficulties, persistent drooling, speech abnormalities, dental malocclusion or psychosocial impact (R34, A+++) [2]. Surgery should be performed by an experienced team after multidisciplinary assessment (paediatric anaesthesia, intensive care, surgery, speech and language therapy, orthodontics), preferably in a reference centre, with outcome audit and follow-up until 16 years (R37–R38, A+++); feeding difficulties require support from a feeding and nutrition specialist (R36, A+++) [2]. GeneReviews adds long teats, tube feeding in severe cases, intubation at birth when needed and tracheostomy for severe respiratory failure with upper airway obstruction [1].

The most widely used technique is anterior wedge resection, among others (V-shaped, W-shaped, stellate, keyhole, central or peripheral), aiming to reduce volume while preserving shape and function; complications are uncommon (oedema and dehiscence) and long-term outcomes are generally favourable for drooling, feeding and speech, without substantial alteration of taste [2, 13]. The Italian Association reports that early intervention was associated with more postoperative complications, that glossectomy does not alter the rate of mandibular growth and that it is not recommended solely to prevent class III malocclusion [13]. Two recent series illustrate practice variability: in a 30-year Canadian review (28 patients with a mean age of 14.3 years, without airway obstruction or feeding problems), no conservatively managed patient required tongue reduction, 7.1% underwent orthognathic surgery, 57% retained class III malocclusion and 80% articulation difficulties, particularly interdental lisp; in a Brazilian series from Curitiba (29 patients, 26 partial glossectomies between 2015 and 2025, median surgical age 10 months), mean reduction was 31.6% in length and 21% in width, with a low complication rate [50, 51]. Resected tissue is useful for molecular testing when blood is negative [1].

7.6 Abdominal wall

Omphalocele treatment follows general recommendations, but in BWSp attention should be paid to hypoglycaemia risk and the anaesthetic risk of severe macroglossia (R39, A+++) [2].

7.7 Growth, lateralised overgrowth and limb discrepancy

Internationally validated syndrome-specific growth charts are lacking (R25, A+++); growth and puberty should be monitored at least annually until growth ends, recognising that final height may exceed +2 standard deviations (R26, A++), and tall stature can be treated as in other children (R27, A++) [2]. Limb discrepancy should be assessed clinically at least once a year during childhood, with referral to paediatric orthopaedics if present (R28–R29) [2]. For discrepancies below 2 cm, a shoe lift may suffice; above 2 cm (predicted at the end of growth), epiphysiodesis is usually indicated, preferably reversible; lengthening of the shorter limb is reserved for specific cases and surgical correction of upper limb asymmetry is generally not indicated (R30–R32) [1, 2]. The Italian Association prefers arresting growth of the longer bone, given the usually tall stature, describes temporary proximal tibial plate epiphysiodesis around 8 years followed by distal femoral epiphysiodesis 2–3 years later, with hardware removal after the pubertal growth spurt, and recommends assessing bone age before surgery; the Ilizarov method is not indicated at preschool or school age [13]. Facial hemihyperplasia should be assessed in a craniofacial centre [1].

7.8 Heart, nervous system and kidney

Cardiovascular examination is recommended at diagnosis in all patients, with echocardiography if there is suspicion, and congenital heart defects are treated as in the general population (R44–R47) [2]. In patients with genomic rearrangements involving IC2, annual electrocardiographic assessment is recommended because of the risk of long QT syndrome and sudden death (R46, B+) [2]. Cognitive development should be followed by the paediatrician, with attention to prematurity, neonatal hypoglycaemia and carriers of rearrangements or genome-wide paternal UPD (R48); brain magnetic resonance imaging is reserved for children with neurological symptoms (R50) [2]. For the kidney, recommendations are screening for nephrourological malformations at diagnosis (R51), attention to hypercalciuria (R52), referral of detected abnormalities to paediatric nephrology and urology (R53), seeking nephrocalcinosis and stones on surveillance ultrasound (R54) and nephrourological assessment with blood pressure and ultrasound at transition to adulthood (R55) [2]. GeneReviews suggests blood pressure and urinary calcium/creatinine ratio annually or biennially and annual renal ultrasound from 8 years to mid-adolescence and periodically in adults [1].

7.9 Tumour surveillance: rationale and guideline divergences

Tumour surveillance aims to detect tumours at an earlier stage, with a lower treatment burden, recognising that no protocol detects all tumours and that surveillance entails costs, false positives and psychological impact [2]. Wilms tumour doubling time has been estimated at 11–13 days, supporting 3–4-month intervals; overall Wilms tumour survival is at least 90% at 4 years, so the expected benefit is primarily reduced treatment morbidity and the possibility of partial nephrectomy [2, 32]. An analysis combining literature cases and two cohorts showed that patients with BWS and screen-detected Wilms tumour had lower stages and that screening until 7 years detects close to 95% of Wilms tumours [52]. For hepatoblastoma, preliminary data suggested an earlier stage and better prognosis with serial αFP measurement, secreted by more than 95% of these tumours, and comparison of presentation age with a control cohort showed that halving screening duration should not compromise efficacy [1, 2, 53]. Neither strategy has been assessed in a randomised trial [2, 32].

The central divergence concerns the risk threshold: North America recommends surveillance from 1% risk and many European countries from 5%; AACR 2017 cites 2% as the usual European threshold, whereas AACR 2024, the consensus and the ACMG cite 5% [2, 10–12]. The consensus attributes the difference to medical and medicolegal cultures and accepts that protocols may differ between Europe and North America provided outcomes are audited [2]. A second argument has come to weigh against stratification: the blood epigenotype may differ from that of the kidney or liver in which the tumour arises, and mosaicism levels vary between tissues within a patient, leading AACR in 2024 and the Italian Association in 2025 to recommend hepatoblastoma surveillance irrespective of genotype [11, 13].

The European SIOP-Europe and SIOP Renal Tumour Study Group document (2021) updated Wilms tumour surveillance recommendations for at-risk children and calls for registration of all patients with predisposition syndromes; the full text was not read for this article, so its specific BWS recommendations are not reproduced [54]. The Italian Association observes that the 2018 IC2 and αFP recommendations had low voting strength and were not universally accepted, and that many Italian and European centres in practice follow schedules close to North American ones [13].

αFP interpretation. In the normal newborn, serum αFP is very high (frequently several thousand kU/L), declines with a half-life of 5.5 days, reaches a mean of 70 kU/L (standard deviation 56) at 4 months and adult values (<10 kU/L) at 7–9 months [5]. In BWSp, values are higher and fall more slowly in the first year; an analysis of 1372 measurements in 147 patients established syndrome-specific reference intervals, with separate values for premature infants and normalisation (<10 ng/mL) at 14 months, without evident subgroup differences [19, 55]. The AACR recommends interpreting the trend rather than an isolated value, not reacting to small rises within the range (intercurrent illness, teething), repeating measurement 6 weeks after rises exceeding 50–100 ng/mL with review of the latest ultrasound, requesting imaging (magnetic resonance imaging or contrast-enhanced ultrasound) with two successive rises and, with rises exceeding 1000 ng/mL, confirming and proceeding directly to imaging [10, 11]. GeneReviews suggests, if αFP rises without an imaging lesion, repeating it with liver function tests 4–6 weeks later and searching exhaustively for a tumour if the value does not fall [1].

Imaging of detected lesions. For hepatic lesions, the principal differential diagnosis is infantile haemangioma, more frequent in BWS; contrast-enhanced ultrasound can distinguish haemangioma from hepatoblastoma without sedation, and magnetic resonance imaging with hepatobiliary contrast is the alternative [10, 11]. In a retrospective study of 12 patients with BWS and hepatic lesions assessed by both techniques (3 hepatoblastomas and 9 benign lesions), contrast-enhanced ultrasound had 67% sensitivity and 89% specificity, magnetic resonance imaging 75% and 100%, and 83% of magnetic resonance examinations required sedation, compared with none of the ultrasound examinations [56]. Nephroblastomatosis is difficult to distinguish from Wilms tumour on conventional imaging [2].

Other tumours. Adrenal carcinoma screening (clinical examination, ultrasound and dehydroepiandrosterone sulphate every 4–6 months) has been proposed, but the tumour is rare and there are no utility data [2]. Neuroblastoma detection using urinary markers in population programmes had minimal impact on morbidity and mortality, and there is no evidence of benefit in BWSp [2, 13].

Tumour treatment. Children with BWSp and Wilms tumour have less metastatic disease, less anaplastic histology and more bilateral synchronous or metachronous tumours; survival is similar to that for sporadic tumours and nephron-sparing surgery is preferred, with outcomes comparable to total nephrectomy in a 2016 analysis [2, 10]. Treatment should be discussed with cooperative groups unless protocols already include specific BWSp recommendations (R67, A+++) [2]. After a tumour, children return to the previous surveillance schedule when post-tumour surveillance ends [10].

Families' perspective. In a North American survey, most parents viewed surveillance as a way to reduce worry rather than a burden, including parents of children with previous false positives; the AACR and ACMG recommend shared decision-making, particularly when risk is low [11, 12]. The Italian Association recognises that universal surveillance may reduce adherence through emotional and logistical burden [13].

7.10 Clinical trials and research

The ClinicalTrials.gov search on 6 October 2026 identified no trials of BWS-targeted therapy; registered studies are observational or diagnostic, such as the Italian maternal genes and epimutations study (NCT06346418, recruiting) and the French RaDiCo-IDMet imprinting disorders cohort (NCT05945576) [57]. Medicines in development for congenital hyperinsulinism generally are relevant to the subgroup with severe hyperinsulinism, although trials do not specify BWS patients.

The sunRIZE results were disclosed only through company announcements, without a peer-reviewed publication at the time of consultation; a negative primary outcome with "clinically meaningful improvements" in secondary and post hoc analyses does not demonstrate efficacy, and recommendation of a presubmission meeting does not equate to a favourable opinion or authorisation [59, 60]. None of these medicines has been studied specifically in BWS, and response may differ according to molecular mechanism and isodisomy extent [22].

7.11 Developments in the past 12 months

The literature search from 1 October 2025 to 6 October 2026 retrieved 74 records. Peer-reviewed publications with practical implications include: the ACMG clinical practice resource on isolated lateralised overgrowth (October 2025), revising the 2009 ACMG recommendation based on a presumed risk above 5%, and establishing affected-tissue testing and shared decision-making [12]; the Italian Association update (October 2025), proposing ART and twinning as suggestive signs, a prenatal score and hepatoblastoma surveillance in all subgroups [13]; the cohort of 85 children with BWS and hyperinsulinism, relating severity to subgroup and isodisomy extension to the KATP region and proposing a genetic algorithm (June 2026) [22]; the Spanish MLID cohort with methylation arrays (April 2026) [25]; the European series of 646 prenatal tests (August 2026) [44]; the Dutch cohort of 46 patients with upd(11)pat (August 2026) [27]; the progression of limb asymmetry in 142 children (December 2025) [34]; the comparison of contrast-enhanced ultrasound and magnetic resonance imaging for hepatic lesions (September 2026) [56]; reports of breast tumours in adolescent and adult women (October 2025) [40]; the macroglossia series with conservative and surgical management [50, 51]; and the case of CDKN1C undetected by exome but identified by genome, described in section 10 [30]. No new authorisation applications, opinions or authorisations for medicines with a BWS indication were identified; in congenital hyperinsulinism, the only regulatory event identified was the 5 October 2026 ersodetug announcement, which does not constitute a regulatory decision [46, 60].

8. Reproductive Options and Family Counselling

8.1 Principles

Recurrence risk depends on the mechanism identified in the proband, its genetic or epigenetic nature and the sex of the transmitting parent; the consensus therefore recommends counselling by professionals experienced in imprinting disorders, taking into account the size, location and parental origin of alterations (R16–R17, A+++) [2]. Approximately 85% of cases are sporadic and 10–15% familial; in familial cases, inheritance is autosomal dominant with parent-of-origin-dependent expression, and most result from CDKN1C variants, chromosomal 11p15 alterations or genetic IC1 alterations [2, 5]. In most families, risk is below 1%; in recurrent families, maternal CDKN1C variants account for approximately 40% and paternal or maternal 11p15 CNVs for approximately 9% of alterations [1]. The Oxford Desk Reference, predating widespread methylation testing, estimated a risk of approximately 5% with a normal karyotype and UPD and no family history, refinable by MLPA [5].

Sources: GeneReviews (Tables 11 and 12) and consensus (Table 2) [1, 2].

8.2 Testing of parents and relatives

Assessment of the parents of a child without a family history includes a medical history focused on childhood BWS manifestations, childhood photographs and ear examination; anterior earlobe creases are relatively common in the population, but posterior helical pits are rarely found outside BWS [1]. As expressivity diminishes with age, parental birth weights and neonatal complications (umbilical hernia, hypoglycaemia) should be sought [5]. Parental testing is guided by the proband's result: karyotyping of both if there is a visible rearrangement; SNP microarray in both if there is a CNV; targeted maternal testing for the CDKN1C variant; no testing with upd(11)pat because of its postzygotic mechanism; review of maternal reproductive history with IC2 LOM or IC1 GOM [1]. If the proband has MLID and the mother has recurrent pregnancy losses, hydatidiform moles or infertility, maternal-effect genes should be considered; pathogenic variants in these genes increase the risk of pre-eclampsia, pregnancy loss, hydatidiform mole and children with imprinting disorders [1, 13]. The Italian Association recommends MLID testing in patients with IC2 LOM when MS-MLPA shows more than one altered DMR, when there is a maternal history of recurrent miscarriages or infertility, monozygotic twinning or recurrence, or clinical signs of another imprinting disorder, and, if MLID is confirmed, testing maternal subcortical complex genes in the mother [13].

Newborn siblings of a proband should be assessed early: genetic testing if there is a known familial alteration and hypoglycaemia monitoring even without prenatal signs [1]. The AACR recommends that siblings undergo clinical assessment if they have BWSp signs and be tested only when the proband has a hereditary cause [11]. Other relatives are at risk only if a parent carries a genetic alteration or rearrangement [1]. DNA banking should be considered in probands without a molecular diagnosis [1]. Young adults with BWSp should be informed of the availability of genetic counselling before starting a family (R69, A+++), and counselling should also cover asymptomatic carriers, such as a woman who inherited a balanced 11p15 translocation from her father [1, 2].

8.3 Assisted reproductive technologies

The association between ART and BWS was suggested in the early 2000s, with ART births 4–6 times more frequent among children with BWS; the Oxford Desk Reference cites a 4.2-fold risk increase after in vitro fertilisation or intracytoplasmic sperm injection, an estimated risk of 1 in 4000 after in vitro fertilisation in an Australian study and no excess cases in a large Danish registry [2, 5, 61]. In Piedmont, between 2005 and 2014, BWS birth prevalence was 1:1126 in ART-conceived children and 1:12,254 in naturally conceived children, with a relative risk of 10.7 (95% confidence interval 4.7–24.2); between 1997 and 2014, 9 of 67 patients (13.4%) were conceived through ART [23]. In EUROCAT, 7.2% of patients (13/181) were conceived through ART, three times the general population rate in participating countries [8]. The consensus concludes that the association is established, but absolute risk is very low, no greater than 1 in 1000 (R6, A+++) [2]. More than 90% of ART-conceived children with BWS have IC2 LOM, compared with approximately 50% of the others; Smith's reports that BWS due to IC2 LOM is ten times more frequent after ART [2, 19]. In Piedmont, of 8 ART-conceived patients tested, 4 had IC2 LOM, 2 upd(11)pat and 2 negative results, leading the authors to suggest that ART may be implicated in mechanisms other than methylation alterations [23]. Subfertility itself, ovarian stimulation and embryo culture are possible factors, and pregnancy achieved after prolonged subfertility is also associated with imprinting disorders [1, 2]. Some series suggest milder phenotypes in ART-associated cases, underpinning the Italian proposal to score ART [13]. Counselling couples using ART should present relative and absolute risk together.

8.4 Twinning

Twinning is much more frequent in BWS; pairs are mostly monozygotic, female and discordant, with IC2 LOM [2, 19]. In EUROCAT, twin pregnancies accounted for 5.7% of cases, three to four times the rate in participating countries [8]. Because of shared fetal circulation, blood or saliva from the unaffected twin may show the methylation alteration, whereas non-haematological samples, such as buccal swabs, are concordant with phenotype; buccal swabs are the preferred diagnostic sample in these cases [1, 2]. Male pairs are less frequent and show other mechanisms, including upd(11)pat and IC1 GOM [1]. In a multiple-pregnancy cohort, the combined score of the two monochorionic twins remained constant despite discordance, suggesting an asymmetrical distribution of a finite number of epigenetically altered cells; the authors proposed assessing all children from an affected monozygotic pregnancy [62]. GeneReviews recommends examining and testing the apparently unaffected twin, preferably in several tissues, and giving serious consideration to tumour surveillance, and a 2025 Italian series recommends testing both twins even without a phenotype [1, 63]. Recurrence risk for siblings of discordant twins is unknown [1].

8.5 Reproductive options and pregnancy surveillance

In families with a known genomic alteration (CDKN1C variant, CNV or rearrangement), prenatal diagnosis by chorionic villus sampling or amniocentesis and preimplantation genetic diagnosis are available; preimplantation testing does not apply to methylation [1, 5]. In all at-risk pregnancies, with or without a known mechanism, maternal serum αFP may be elevated at 15–16 weeks if omphalocele is present, and ultrasound at 19–20 weeks and again at 25–32 weeks assesses growth (which becomes excessive at the end of the second trimester), abdominal wall defects, organomegaly, renal abnormalities, cleft palate, cardiac defects and macroglossia; the Oxford Desk Reference proposes starting with a dating scan at 8–10 weeks [1, 5]. A normal ultrasound does not eliminate residual risk, and the newborn should be monitored for hypoglycaemia even without prenatal findings [1]. Limitations of prenatal methylation testing (mosaicism, tissue discordance, false positives in villi, false negatives) should be discussed before sampling, together with the ethical issues involved (R20) [2, 44]. The ideal time for counselling and testing at-risk relatives is before pregnancy [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1 Frequency

Frequency estimates depend on definition (classical BWS or spectrum), ascertainment and period, and all tend to underestimate the spectrum by excluding mild cases [1, 10].

The difference between the Italian value (approximately 10 per 100,000) and the European EUROCAT value (3.5–3.8 per 100,000) primarily reflects the difference between a rare disease registry with active clinical and molecular confirmation and a congenital anomaly registry focused on classical BWS, and should not be interpreted as a biological difference between countries [6, 8]. The same caution applies to Spanish and Japanese values [3].

9.2 Ethnic and population diversity

Smith's reports that Caucasian patients have a higher frequency of omphalocele, as in the general paediatric population, while umbilical hernias are more frequent in African-American patients; non-Caucasian patients had more frequent nephromegaly and hyperinsulinism, sometimes requiring partial pancreatectomy [19]. The Philadelphia study underlying these data compared Caucasian, mixed and non-Caucasian groups and concluded that manifestations and molecular diagnoses differ between groups, suggesting ethnic effects on genotype–phenotype correlations [64]. Validation of the scoring system showed variability between ethnic groups, and GeneReviews notes that mechanism frequencies vary between populations [1, 13]. No founder effects were described in the sources consulted, consistent with the predominance of sporadic postzygotic epimutations.

9.3 Sex, twinning and ART as population modifiers

MLID is more frequent in females (4:1) and discordant monozygotic twin pairs are predominantly female [1, 2]. In EUROCAT, macrosomia affected 41.1% of term boys and 24% of term girls [8]. The proportion of ART-conceived cases (7.2% in EUROCAT, 13.4% in Piedmont, up to 15% according to the Italian Association) depends on the use of these technologies in each country [8, 13, 23].

9.4 Brazil and Portugal

Among the sources consulted, the only Brazilian series is from a craniofacial reference centre in Curitiba, analysing records of 29 patients with BWS and macroglossia and 26 partial glossectomies performed between 2015 and 2025; the abstract reports 65.4% female patients and proposes a clinical macroglossia classification (mild in 11.5%, moderate in 46.2% and severe in 42.3%), percentages corresponding to the denominator of 26 operated patients rather than the total of 29 [51]. This is a single-centre surgical series, which cannot estimate frequency or mechanism distribution. No Portuguese data were found in the sources consulted; this absence does not demonstrate a lack of publications.

9.5 Reference networks and patient associations

In Europe, BWS is included in thematic group MTG5 (growth and genetic obesity syndromes) of Endo-ERN, the European reference network for rare endocrine conditions, with more than 100 centres in 27 Member States and Norway, listing Prader-Willi syndrome, Silver-Russell syndrome and BWS among the imprinting disorders within its remit [16]. Endo-ERN disseminates the 2024 joint statement on MLID [16, 26]. The 2018 consensus was organised by the European Network for Congenital Imprinting Disorders (EUCID.net, www.imprinting-disorders.eu) [2, 17]. The consensus recommends providing families with patient association contacts after diagnosis (R72) [2]. Associations mentioned in the sources include the Beckwith-Wiedemann Children's Foundation International, Associazione Italiana Sindrome di Beckwith-Wiedemann (AIBWS) and the United Kingdom Beckwith-Wiedemann Syndrome Support Group [1, 5].

10. Illustrative Clinical Case

10.1 Familial BWS due to a maternal CDKN1C variant undetected by exome sequencing (Wu and colleagues, 2026)

A female newborn, the second daughter of a 27-year-old mother (gravida 3, para 1) without BWS signs, was delivered by caesarean section at 35 weeks and 2 days because of severe pre-eclampsia (blood pressure 157/97 mmHg, proteinuria 2+) [30]. Ultrasound at 20 weeks had shown omphalocele, and amniocentesis performed at another hospital revealed a 46,XX karyotype and normal microarray, without 11p15 testing [30]. At birth she weighed 3450 g (above +2 standard deviations for gestational age) and had macroglossia, bilateral earlobe creases, facial nevus simplex, hepatomegaly and a 3 × 2 cm abdominal wall defect, corrected immediately; echocardiography showed a patent foramen ovale and transfontanelle ultrasound a left subependymal haemorrhage [30]. The photograph of neonatal ear creases corresponds to Figure 1 of the article (Figure 1).

Blood 11p15.5 MS-MLPA was normal (methylation and copy number) and exome sequencing identified no relevant variants; genome sequencing detected a 19-nucleotide duplication in CDKN1C exon 1, NM_000076.2:c.617_635dup p.(Ala213GlyfsTer34), confirmed by Sanger sequencing in the proband and mother (heterozygous) and absent in the father and sister, classified as pathogenic (criteria PVS1, PM2, PP1 and PP4) and submitted to ClinVar (SCV006303522) [30]. The variant lies in the proline–alanine repeat domain and is predicted to eliminate the C-terminal PCNA-binding QT domain (Figure 2, showing the pedigree and electropherograms) [30]. In the MANE Select transcript, the same variant corresponds to NM_001122630.2:c.584_602dup p.(Ala202GlyfsTer34), and at genomic level to NC_000011.10:g.2884858_2884876dup (GRCh38), according to VariantValidator [31]. The authors attribute exome failure to GC richness and exon 1 repeats, which reduce capture, but acknowledge that they did not analyse read depth to confirm this [30].

Follow-up over 54 months showed mild gross motor and expressive language delays at 12 months (Bayley-III), macroglossia with persistent protrusion (Figure 3), reduction glossectomy at 3 years for articulation and swallowing difficulties, normalisation of development at 3 years, weight consistently above the 97th centile and a limb discrepancy of 0.5 cm at 12 months that disappeared by 54 months [30]. αFP was 13.26 IU/mL at 12 months, 8.78 at 15, 16.80 at 36, 2.12 at 48 and 1.90 at 54 months, with no tumours detected (Figure 4) [30].

The case illustrates several points of this article. First, prenatal omphalocele with a normal karyotype and microarray does not exclude BWS, and the consensus algorithm, providing for CDKN1C after negative methylation testing with an abdominal wall defect, was followed only postnatally [2, 30]. Second, a negative exome result does not exclude a CDKN1C variant; the authors suggest optimised Sanger sequencing or genome sequencing [30]. Third, the asymptomatic carrier mother, whom the authors presume inherited the variant from her father (an unconfirmed hypothesis because a grandfather sample was unavailable), demonstrates silent paternal transmission and the 50% risk for each subsequent child of this mother [1, 30]. Fourth, the CDKN1C subgroup has its own tumour profile: the authors continued αFP follow-up and planned a neuroblastoma protocol with abdominal ultrasound and urinary vanillylmandelic and homovanillic acids every three months until 6 years, with chest radiographs, then every six months until 10 years [30]. This plan corresponds to the proposal described in GeneReviews and AACR 2017; the Italian Association advises against routine chest radiography and the 2018 consensus is limited to abdominal ultrasound until 7 years, so the report's management should not be taken as a general recommendation [1, 2, 10, 13]. Transient αFP fluctuations without a tumour are consistent with the AACR 2024 recommendation not to measure αFP in this subgroup [11].

11. Prognosis and Long-Term Follow-up

11.1 Prognosis

After the neonatal period and early years, the prognosis for health and quality of life is favourable: growth slows, macroglossia and facial features diminish and most adults are healthy [1, 5, 19]. Morbidity and mortality are concentrated in the neonatal period (prematurity, airway obstruction, hypoglycaemia) and early years (tumours) [1, 8]. Children with BWSp-associated Wilms tumour have overall survival similar to sporadic tumours (at least 90% at 4 years), and early diagnosis through surveillance allows smaller tumours and nephron-sparing surgery [2]. Neurological prognosis depends primarily on preventing severe hypoglycaemia and prematurity complications, and on unbalanced chromosomal alterations or genome-wide UPD when present [1, 2].

Adult data are scarce. In an Italian series of 34 adults (18–58 years), 44% had tall final stature, 85% had required BWS-related surgery and sequelae included pronunciation or swallowing difficulties due to macroglossia (9), painful scoliosis due to lateralised overgrowth (4), recurrent urinary stones (4), azoospermia probably consequent on cryptorchidism (4), severe intellectual disability attributed to neonatal asphyxia (2) and diabetes after subtotal pancreatectomy (1); diagnoses included hepatoblastoma at 22 years, lymphoblastic leukaemia at 21 and an adrenal adenoma and Sertoli cell tumour at 22 and 24 years, as well as benign tumours [39]. The consensus found no evidence of an increased adult carcinoma risk warranting surveillance, the Italian Association considers that the approximately 25% estimated adult tumour risk probably results from study design and recommends only a low threshold for investigating symptoms, and four adults in the Italian series had healthy children [2, 13, 39]. Male fertility appears reduced, without a known infertility frequency [2].

11.2 Childhood follow-up

11.3 Transition and adulthood

The consensus recommends review at 16–18 years to identify complications requiring adult medical follow-up, with detailed clinical review and renal ultrasound at 16 years, nephrourological assessment with blood pressure at transition and subsequent follow-up directed only at existing problems (R55, R68) [2]. Patients with congenital heart disease require adult clinic follow-up, and carriers of IC2 CNVs or rearrangements with long QT risk should be followed lifelong [2]. The Italian Association reproduces Greer and colleagues' adult proposals (counselling and male fertility assessment when appropriate, echocardiography and renal function every 3–5 years, renal ultrasound once in early adulthood, hearing every 2–3 years), and the consensus notes that, without childhood abnormalities, the yield of these tests in asymptomatic adults is probably low [2, 13]. Transition preparation includes educating adolescents about their disease and addressing paediatric sequelae such as macroglossia, scoliosis, limb discrepancy, stones, cryptorchidism and urinary infections [13].

11.4 Psychosocial support

Diagnosis, repeated surveillance and concerns about tumour risk and appearance (macroglossia, drooling) may affect the well-being of children and families; emotional and peer relationship difficulties have been described, and recommendation divergences generate parental uncertainty [2, 13]. Professionals should adopt a holistic approach and offer referral for specialist counselling, psychology and social services (R71, A+++) [2]. Parent questionnaires have been proposed to screen for emotional and behavioural problems without increasing the care burden [13].

12. Summary and Final Considerations

Beckwith-Wiedemann spectrum results from dysregulation of two imprinted gene domains at 11p15.5, controlled by IC1 (H19/IGF2:IG-DMR) and IC2 (KCNQ1OT1:TSS-DMR), through loss of maternal IC2 methylation (approximately 50%), mosaic segmental paternal uniparental disomy (approximately 20%), gain of IC1 methylation (5–10%), maternal CDKN1C variants (5% of sporadic and 40% of familial cases) or chromosomal alterations, with approximately 20% remaining without a molecular diagnosis [1, 2]. The 2018 international consensus replaced previous criteria with a score of cardinal and suggestive signs: 4 points establish clinical diagnosis of classical BWS and 2 points indicate molecular testing, beginning with simultaneous IC1 and IC2 methylation with copy number [2].

Mosaicism shapes the entire pathway: it explains lateralised overgrowth, discordance between twins and tissues, possible negative blood tests and the need to analyse other tissues, and limits the ability of the blood epigenotype to predict tumour risk in each organ [2, 11, 12]. Overall tumour risk is approximately 8%, ranges from 2.6% in IC2 to 28% in IC1, concentrates Wilms tumour in the first 7 years and hepatoblastoma in the first 30 months, and the CDKN1C subgroup is associated with neuroblastoma rather than Wilms tumours or hepatoblastomas [2, 9, 36]. Guidelines diverge: the European consensus dispenses with ultrasound in IC2 and with αFP, AACR (2024) recommends complete abdominal ultrasound and αFP every three months until 3 years in all subgroups except CDKN1C, followed by renal ultrasound every three months until 7 years, a schedule adopted by the ACMG (2025) in isolated lateralised overgrowth when surveillance is indicated, and the Italian Association (2025) moved closer to this position for hepatoblastoma [2, 11–13]. Protocol choice should be explained to the family and consistent with the healthcare system in which the child is followed.

Hyperinsulinism is the neonatal emergency of the disease; blood glucose monitoring in the first 48 hours, fasting testing before discharge and knowledge of molecular subgroup, particularly paternal isodisomy extension to the ABCC8/KCNJ11 region, guide treatment with diazoxide, somatostatin analogues or pancreatectomy [2, 22]. Genetic counselling depends on mechanism: risk generally below 1% with epimutations and UPD, up to 50% with maternally transmitted CDKN1C variants and with CNVs, IC1 SNVs and rearrangements, and high risk when MLID reflects a maternal variant in a maternal-effect gene [1, 2]. The association with assisted reproductive technologies is established, with a relative risk of approximately 10 and an absolute risk no greater than 1 in 1000 [2, 23]. The true tumour risk of isolated lateralised overgrowth, the clinical utility of MLID detection, adult natural history and the comparative benefit of European and North American surveillance strategies remain unresolved, and can only be measured through prospective international registries [2, 12, 26].

13. References

1. Shuman C, Kalish JM, Weksberg R. Beckwith-Wiedemann Syndrome. 2000 Mar 3 [updated 2023 Jun 8; reviewed 2023 Sep 21]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. NBK1394. PMID: 20301568. [https://www.ncbi.nlm.nih.gov/books/NBK1394/](https://www.ncbi.nlm.nih.gov/books/NBK1394/)
2. Brioude F, Kalish JM, Mussa A, Foster AC, Bliek J, Ferrero GB, et al. Expert consensus document: Clinical and molecular diagnosis, screening and management of Beckwith-Wiedemann syndrome: an international consensus statement. Nat Rev Endocrinol. 2018;14(4):229-49. doi:10.1038/nrendo.2017.166. PMID: 29377879. PMC6022848.
3. Orphanet. Beckwith-Wiedemann syndrome (ORPHA:116) and aetiological subtypes (ORPHA:231117, 96193, 231120, 231127, 96076, 231130). Orphadata: epidemiology and cross-references (version dated 23 Jun 2026). [https://www.orpha.net/en/disease/detail/116](https://www.orpha.net/en/disease/detail/116). Accessed 6 Oct 2026.
4. GeneGuidance. Gene records for CDKN1C, KCNQ1OT1, H19, IGF2 and KCNQ1 (clinical curation and identifiers). User database, version dated 26 Sep 2026. Accessed 6 Oct 2026.
5. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Beckwith-Wiedemann syndrome (BWS); p. 358–360.
6. Mussa A, Russo S, De Crescenzo A, Chiesa N, Molinatto C, Selicorni A, et al. Prevalence of Beckwith-Wiedemann syndrome in North West of Italy. Am J Med Genet A. 2013;161A(10):2481-6. doi:10.1002/ajmg.a.36080. PMID: 23918458.
7. Thorburn MJ, Wright ES, Miller CG, Smith-Read EH. Exomphalos-macroglossia-gigantism syndrome in Jamaican infants. Am J Dis Child. 1970;119(4):316-21. doi:10.1001/archpedi.1970.02100050318006. PMID: 5434588.
8. Barisic I, Boban L, Akhmedzhanova D, Bergman JEH, Cavero-Carbonell C, Grinfelde I, et al. Beckwith Wiedemann syndrome: A population-based study on prevalence, prenatal diagnosis, associated anomalies and survival in Europe. Eur J Med Genet. 2018;61(9):499-507. doi:10.1016/j.ejmg.2018.05.014. PMID: 29753922.
9. Maas SM, Vansenne F, Kadouch DJ, Ibrahim A, Bliek J, Hopman S, et al. Phenotype, cancer risk, and surveillance in Beckwith-Wiedemann syndrome depending on molecular genetic subgroups. Am J Med Genet A. 2016;170(9):2248-60. doi:10.1002/ajmg.a.37801. PMID: 27419809.
10. Kalish JM, Doros L, Helman LJ, Hennekam RC, Kuiper RP, Maas SM, et al. Surveillance recommendations for children with overgrowth syndromes and predisposition to Wilms tumors and hepatoblastoma. Clin Cancer Res. 2017;23(13):e115-22. doi:10.1158/1078-0432.CCR-17-0710. PMID: 28674120. PMC5538793.
11. Kalish JM, Becktell KD, Bougeard G, Brodeur GM, Diller LR, Doria AS, et al. Update on surveillance for Wilms tumor and hepatoblastoma in Beckwith-Wiedemann syndrome and other predisposition syndromes. Clin Cancer Res. 2024;30(23):5260-9. doi:10.1158/1078-0432.CCR-24-2100. PMID: 39320341. PMC11611621.
12. Erwin AL, El Haija AA, Bennett JT, Biesecker LG, Deardorff MA, Hathaway ER, et al. Isolated lateralized overgrowth and the need for tumor screening: A clinical practice resource of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2025;27(10):101480. doi:10.1016/j.gim.2025.101480. PMID: 40693985. PMC12313104.
13. Russo S, Milani D, Meossi C, Marcucci L, Pajno R, Butti N, et al. Beckwith-Wiedemann spectrum (BWSp): an update on diagnosis, management, and follow-up from the scientific committee of the Italian BWSp association. Ital J Pediatr. 2025;51(1):287. doi:10.1186/s13052-025-02131-3. PMID: 41126215. PMC12542140.
14. Teva Pharmaceuticals USA. PROGLYCEM (diazoxide) oral suspension. Prescribing information, rev. 7/2024. NDA 017453. DailyMed set ID b16c7832-2fd9-49af-b923-1dc0d91fd6e2. [https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b16c7832-2fd9-49af-b923-1dc0d91fd6e2](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b16c7832-2fd9-49af-b923-1dc0d91fd6e2). Accessed via openFDA on 6 Oct 2026.
15. Novartis. SANDOSTATIN (octreotide acetate) injection; SANDOSTATIN LAR Depot. Prescribing information. NDA 019667 and NDA 021008. DailyMed set ID 4e2c9856-1836-49f0-9472-4dbeeb408f39. [https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=4e2c9856-1836-49f0-9472-4dbeeb408f39](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=4e2c9856-1836-49f0-9472-4dbeeb408f39). Accessed via openFDA on 6 Oct 2026.
16. European Reference Network on Rare Endocrine Conditions (Endo-ERN). Growth & Genetic Obesity Syndromes (MTG5). [https://endo-ern.eu/rare-growth-general-obesity-syndromes/](https://endo-ern.eu/rare-growth-general-obesity-syndromes/). Accessed 6 Oct 2026.
17. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Genomic imprinting; p. 30–31.
18. Wiedemann HR. Complexe malformatif familial avec hernie ombilicale et macroglossie — un «syndrome nouveau»? J Genet Hum. 1964;13:223-32. PMID: 14231762.
19. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Beckwith-Wiedemann syndrome; p. 218–223.
20. Weksberg R, Shen DR, Fei YL, Song QL, Squire J. Disruption of insulin-like growth factor 2 imprinting in Beckwith-Wiedemann syndrome. Nat Genet. 1993;5(2):143-50. doi:10.1038/ng1093-143. PMID: 8252039.
21. Weksberg R, Teshima I, Williams BR, Greenberg CR, Pueschel SM, Chernos JE, et al. Molecular characterization of cytogenetic alterations associated with the Beckwith-Wiedemann syndrome (BWS) phenotype refines the localization and suggests the gene for BWS is imprinted. Hum Mol Genet. 1993;2(5):549-56. doi:10.1093/hmg/2.5.549. PMID: 8518793.
22. George AM, Viswanathan A, Sussman JH, Hathaway ER, Casalnova A, Nesbitt S, et al. Determinants of hyperinsulinism severity in children with Beckwith-Wiedemann syndrome. J Clin Endocrinol Metab. 2026;111(7):e1823-31. doi:10.1210/clinem/dgag053. PMID: 41655234. PMC13271782.
23. Mussa A, Molinatto C, Cerrato F, Palumbo O, Carella M, Baldassarre G, et al. Assisted reproductive techniques and risk of Beckwith-Wiedemann syndrome. Pediatrics. 2017;140(1):e20164311. doi:10.1542/peds.2016-4311. PMID: 28634246.
24. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Overgrowth; p. 272–275.
25. Cazalla M, Parra A, Rodríguez-Antolín C, Silván C, Miranda-Alcaraz L, Mora-Gómez M, et al. Investigation of multilocus imprinting disturbance (MLID) in 101 Beckwith-Wiedemann spectrum patients. Clin Epigenetics. 2026;18(1):131. doi:10.1186/s13148-026-02077-1. PMID: 42036661. PMC13339320.
26. Mackay DJG, Gazdagh G, Monk D, Brioude F, Giabicani E, Krzyzewska IM, et al. Multi-locus imprinting disturbance (MLID): interim joint statement for clinical and molecular diagnosis. Clin Epigenetics. 2024;16(1):99. doi:10.1186/s13148-024-01713-y. PMID: 39090763.
27. Maas SM, Lauffer P, Vos N, Lombardi P, Mannens MMAM, Zegers Y, et al. Clinical and molecular characterization of 46 patients with Beckwith-Wiedemann spectrum and uniparental disomy of 11p15. Am J Med Genet A. 2026 Aug 18 [advance online publication]. doi:10.1002/ajmg.a.70281. PMID: 42613680.
28. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Hypoglycaemia in neonate and infant; p. 182–185.
29. Arnoux JB, de Lonlay P. Congenital hyperinsulinism and genetic disorders of insulin resistance and signalling. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 6, sections 6.1–6.6. doi:10.1007/978-3-662-63123-2_6.
30. Wu S, Zhang Y, Zhang H, Yan X, Pi Y. A novel maternally inherited CDKN1C variant in a familial Beckwith-Wiedemann syndrome case: expanding the genotype-phenotype spectrum. BMC Med Genomics. 2026;19(1):70. doi:10.1186/s12920-026-02351-4. PMID: 41888894. PMC13141551.
31. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-8. doi:10.1002/humu.23348. PMID: 28967166. Consultation of NM_000076.2:c.617_635dup (GRCh38, VariantValidator 4.0.1, vvdb_2025_3) at [https://variantvalidator.org](https://variantvalidator.org) on 6 Oct 2026.
32. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Wilms tumour; p. 620–622.
33. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Hemihypertrophy and limb asymmetry; p. 164–166.
34. Lopez RD, Hall CE, Sussman JH, George AM, Phillips CA, Gerace C, et al. Evolution of discrepancies in limb asymmetry in Beckwith-Wiedemann spectrum. J Child Orthop. 2025;19(6):463-72. doi:10.1177/18632521251384575. PMID: 41111602. PMC12528058.
35. Mussa A, Molinatto C, Baldassarre G, Riberi E, Russo S, Larizza L, et al. Cancer risk in Beckwith-Wiedemann syndrome: a systematic review and meta-analysis outlining a novel (epi)genotype specific histotype targeted screening protocol. J Pediatr. 2016;176:142-149.e1. doi:10.1016/j.jpeds.2016.05.038. PMID: 27372391.
36. George AM, Viswanathan A, Best LG, Monahan C, Limmina M, Ganguly A, et al. Expanded phenotype and cancer risk in patients with Beckwith-Wiedemann spectrum caused by CDKN1C variants. Am J Med Genet A. 2024;194(10):e63777. doi:10.1002/ajmg.a.63777. PMID: 38822599.
37. Hoyme HE, Seaver LH, Jones KL, Procopio F, Crooks W, Feingold M. Isolated hemihyperplasia (hemihypertrophy): report of a prospective multicenter study of the incidence of neoplasia and review. Am J Med Genet. 1998;79(4):274-8. PMID: 9781907.
38. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Cardiomyopathy in children under 10 years; p. 85.
39. Gazzin A, Carli D, Sirchia F, Molinatto C, Cardaropoli S, Palumbo G, et al. Phenotype evolution and health issues of adults with Beckwith-Wiedemann syndrome. Am J Med Genet A. 2019;179(9):1691-702. doi:10.1002/ajmg.a.61301. PMID: 31339634.
40. Grossmann LD, George AM, Lapidus T, Ligh C, MacFarland SP, Kalish JM. Breast findings in females with Beckwith-Wiedemann syndrome. Pediatr Blood Cancer. 2026;73(1):e32109. doi:10.1002/pbc.32109. PMID: 41074634.
41. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Anterior abdominal wall defects; p. 704–705.
42. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Urinary tract and renal anomalies (CAKUT); p. 797.
43. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Polyhydramnios; p. 774–775.
44. Connolly M, McClelland L, Tannorella P, Richter T, Mainini E, Dufke A, et al. Prenatal genetic testing for Beckwith-Wiedemann syndrome: considerations, challenges and observations (a real-world study). Prenat Diagn. 2026;46(9):1486-93. doi:10.1002/pd.70206. PMID: 42347556. PMC13446535.
45. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Coarse facial features; p. 106–107.
46. European Medicines Agency. Medicines output — list of human medicines: diazoxide, hyperinsulinism and dasiglucagon; Zegalogue and Viokat EPARs. [https://www.ema.europa.eu/en/medicines/human/EPAR/zegalogue](https://www.ema.europa.eu/en/medicines/human/EPAR/zegalogue). Accessed 6 Oct 2026.
47. Soleno Therapeutics. VYKAT XR (diazoxide choline) extended-release tablets. Prescribing information. NDA 216665. DailyMed set ID 0e745e85-9512-e360-e063-6394a90aebf1. [https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0e745e85-9512-e360-e063-6394a90aebf1](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0e745e85-9512-e360-e063-6394a90aebf1). Accessed via openFDA on 6 Oct 2026.
48. Zealand Pharma. ZEGALOGUE (dasiglucagon) injection. Prescribing information. NDA 214231. DailyMed set ID 14704879-872c-4967-8779-04a3bbdfb4e6. [https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=14704879-872c-4967-8779-04a3bbdfb4e6](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=14704879-872c-4967-8779-04a3bbdfb4e6). Accessed via openFDA on 6 Oct 2026.
49. ClinicalTrials.gov. A two-period, open-label trial evaluating the efficacy and safety of dasiglucagon for the treatment of children with congenital hyperinsulinism. NCT03777176. [https://clinicaltrials.gov/study/NCT03777176](https://clinicaltrials.gov/study/NCT03777176). Accessed 6 Oct 2026.
50. Bouhadana GC, Delisle É, Borsuk DE, Bortoluzzi P. The natural evolution of macroglossia among Beckwith-Wiedemann patients: a 30-year institutional review. J Craniofac Surg. 2026;37(7-8):1963-8. doi:10.1097/SCS.0000000000012659. PMID: 41891532.
51. da Silva Freitas R, Alves LG, Mateus M, de Oliveira IC, Linnenkamp MDW, Nasser IJG, et al. Final tongue size in the treatment of macroglossia associated with Beckwith-Wiedemann syndrome: review of 29 cases. Ann Plast Surg. 2026 Sep 2 [advance online publication]. doi:10.1097/SAP.0000000000004862. PMID: 42684909.
52. Mussa A, Duffy KA, Carli D, Griff JR, Fagiano R, Kupa J, et al. The effectiveness of Wilms tumor screening in Beckwith-Wiedemann spectrum. J Cancer Res Clin Oncol. 2019;145(12):3115-23. doi:10.1007/s00432-019-03038-3. PMID: 31583434. PMC6876630.
53. Mussa A, Duffy KA, Carli D, Ferrero GB, Kalish JM. Defining an optimal time window to screen for hepatoblastoma in children with Beckwith-Wiedemann syndrome. Pediatr Blood Cancer. 2019;66(1):e27492. doi:10.1002/pbc.27492. PMID: 30270492. PMC7955797.
54. Hol JA, Jewell R, Chowdhury T, Duncan C, Nakata K, Oue T, et al. Wilms tumour surveillance in at-risk children: literature review and recommendations from the SIOP-Europe Host Genome Working Group and SIOP Renal Tumour Study Group. Eur J Cancer. 2021;153:51-63. doi:10.1016/j.ejca.2021.05.014. PMID: 34134020.
55. Duffy KA, Cohen JL, Elci OU, Kalish JM. Development of the serum α-fetoprotein reference range in patients with Beckwith-Wiedemann spectrum. J Pediatr. 2019;212:195-200.e2. doi:10.1016/j.jpeds.2019.05.051. PMID: 31235384. PMC6707865.
56. Davis LM, Ntoulia A, Lerebo WT, Kalish JM, Macfarland S, Dennis R, et al. Assessing focal liver lesions in Beckwith-Wiedemann syndrome: comparison of contrast-enhanced ultrasound and magnetic resonance imaging. Ultraschall Med. 2026 Sep 1 [advance online publication]. doi:10.1055/a-2890-9689. PMID: 42679826.
57. ClinicalTrials.gov. Search for "Beckwith-Wiedemann" (API v2); studies NCT06346418 and NCT05945576. [https://clinicaltrials.gov/search?cond=Beckwith-Wiedemann](https://clinicaltrials.gov/search?cond=Beckwith-Wiedemann). Accessed 6 Oct 2026.
58. ClinicalTrials.gov. A Phase 3, randomized, double-blind, placebo-controlled, parallel-arm study to evaluate the efficacy and safety of RZ358 in patients with congenital hyperinsulinism (sunRIZE). NCT06208215. [https://clinicaltrials.gov/study/NCT06208215](https://clinicaltrials.gov/study/NCT06208215). Accessed 6 Oct 2026.
59. Rezolute, Inc. Rezolute provides insights from its Phase 3 sunRIZE study in congenital hyperinsulinism (press release). Form 8-K, Exhibit 99.1, filed with the SEC on 7 Jan 2026. [https://www.sec.gov/Archives/edgar/data/1509261/000110465926001556/tm262357d1_ex99-1.htm](https://www.sec.gov/Archives/edgar/data/1509261/000110465926001556/tm262357d1_ex99-1.htm). Accessed 6 Oct 2026.
60. Rezolute, Inc. Rezolute announces FDA recommendation to hold a pre-BLA meeting as a precursor to submitting a BLA for congenital hyperinsulinism (press release). Form 8-K, Exhibit 99.1, 5 Oct 2026. [https://www.sec.gov/Archives/edgar/data/1509261/000110465926113503/tm2626976d1_ex99-1.htm](https://www.sec.gov/Archives/edgar/data/1509261/000110465926113503/tm2626976d1_ex99-1.htm). Accessed 6 Oct 2026.
61. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Assisted reproductive technology (IVF, ICSI, PGD); p. 706, 708.
62. Cohen JL, Duffy KA, Sajorda BJ, Hathaway ER, Gonzalez-Gandolfi CX, Richards-Yutz J, et al. Diagnosis and management of the phenotypic spectrum of twins with Beckwith-Wiedemann syndrome. Am J Med Genet A. 2019;179(7):1139-47. doi:10.1002/ajmg.a.61164. PMID: 31067005. PMC7959854.
63. Bellani I, Trevisan V, Viscogliosi G, Pomponi MG, Chiurazzi P, Mussa A, et al. Challenges and pitfalls in diagnosing twins with discordant BWS phenotype. Clin Genet. 2025;108(5):609-11. doi:10.1111/cge.70022. PMID: 41054832. PMC12501699.
64. Duffy KA, Sajorda BJ, Yu AC, Hathaway ER, Grand KL, Deardorff MA, et al. Beckwith-Wiedemann syndrome in diverse populations. Am J Med Genet A. 2019;179(4):525-33. doi:10.1002/ajmg.a.61053. PMID: 30719840. PMC6454922.