Genetic Macrocephaly — Approach and Monogenic Forms

Clinical, imaging and genetic approach to the child with increased head circumference, covering PI3K-AKT-mTOR pathway megalencephalies (MCAP, MPPH, hemimegalencephaly and PTEN), overgrowth syndromes due to epigenetic regulators (Sotos, Weaver and Tatton-Brown–Rahman), benign familial macrocephaly with benign enlargement of the subarachnoid spaces and megalencephalic leukodystrophies.

Literature and regulatory information updated: 9 October 2026.

0. Initial Reference Table

Entities covered in this article and identifiers.

Nomenclature note: Sotos syndrome has code ORPHA:821; ORPHA:1627, sometimes associated with NSD1 in aggregated databases, corresponds to 5q35 deletion syndrome [31].

1. Introduction

1.1. Definitions: macrocephaly, macrocrania and megalencephaly

Macrocephaly is an anthropometric finding rather than a diagnosis. It is defined as occipitofrontal head circumference greater than 2 SD above the mean, or above the 97th percentile, for age and sex; this range includes 2–3% of healthy children and up to 5% of the paediatric population, so macrocephaly between +2 and +3 SD often coexists with normal development and, in up to 60% of cases, with familial recurrence of benign macrocephaly, whereas above +3 SD (≈99.7th percentile) neurogenetic disorders with ID, autism spectrum disorder (ASD) and other comorbidities predominate [2]. Sources differ in the threshold adopted: the Oxford Desk Reference: Clinical Genetics and Genomics (OXF) defines macrocephaly as HC more than 3 SD above the mean and recommends comparing it with length or height to determine whether it is proportionate, whereas the same book, in its overgrowth chapter, and the MPPH, Sotos, TBRS and Smith-Kingsmore GeneReviews use ≥2 SD (≈98th percentile) as an inclusion criterion [1, 3–5, 12, 14]. HC >2 SD is adopted as the definition of macrocephaly; the +3 SD threshold serves as a marker of greater likelihood of a pathological cause, not as a definition.

Macrocrania denotes an enlarged skull regardless of contents and is used synonymously with macrocephaly [2]. Megalencephaly is increased size and weight of the brain itself and represents only one cause of macrocephaly; the others are increased intra- or extra-axial cerebrospinal fluid (CSF) (hydrocephalus, enlarged subarachnoid spaces, subdural collections), masses and cysts, and bone thickening in skeletal dysplasias [2]. Megalencephaly is divided into anatomical or developmental, from increased cell number or size due to dysregulated proliferation pathways, and metabolic, from substance accumulation; only developmental megalencephaly may be unilateral or focal and, in that case, may not increase HC [2]. Congenital (primary) macrocephaly is present in utero or at birth; secondary macrocephaly develops after birth; relative macrocephaly corresponds to HC below +2 SD but disproportionate to height and weight, as in skeletal dysplasias with short stature [1, 2]. Overgrowth is regional or global growth above +2 SD; global overgrowth, with increased height and HC, suggests a syndrome when accompanied by ID or a congenital abnormality [5].

1.2. Historical milestones and molecular discovery

Sotos syndrome was described by Sotos and colleagues in 1964 in five children with cerebral gigantism, and NSD1 haploinsufficiency was demonstrated in 2002 by Kurotaki and colleagues, who isolated the gene from a translocation's 5q35 breakpoint and found one nonsense variant, three frameshift variants and 20 submicroscopic microdeletions in 42 sporadic Japanese cases [32, 33]. Douglas and colleagues showed in 2003, in a British series of 75 children with overgrowth, that intragenic variants were the main cause outside Japan (NSD1 alterations in 28 of 37 typical phenotypes) [34]. Weaver syndrome was named after David Weaver, who described two boys with accelerated bone maturation, unusual facial appearance and camptodactyly in 1974; in 2011–2012, two groups identified germline EZH2 variants, one through trio exome sequencing including one of the original families [11, 35, 36]. In 2017, EED and SUZ12, the other Polycomb repressive complex 2 (PRC2) subunits, were added [37]. Tatton-Brown–Rahman syndrome was defined in 2014 from DNMT3A variants in 13 of 152 individuals with overgrowth and ID, versus none in 1000 controls [38]. Malan syndrome, formerly called Sotos type 2, was associated with NFIX by Malan and colleagues in 2010 [39].

In the phosphatidylinositol 3-kinase (PI3K)–AKT–mTOR pathway group, macrocephaly–congenital cutis marmorata telangiectatica was recognised in 1997 by Clayton-Smith and by Moore and colleagues; MPPH was described in the neurological literature as a separate entity, and Gripp and colleagues reported three markedly overlapping cases in 2009 [40]. In 2012, Mirzaa and colleagues reviewed 42 patients, showed that skin lesions are capillary malformations rather than true cutis marmorata, and proposed megalencephaly–capillary malformation (MCAP), with distinct MCAP and MPPH criteria [41]. In the same year, Rivière and colleagues identified de novo germline or postzygotic variants through exome sequencing in AKT3, PIK3R2 (one recurrent variant in 11 MPPH families) and PIK3CA (15 variants, mostly postzygotic, in 23 MCAP patients) [42]. In 2014, discovery of CCND2 variants stabilising cyclin D2 provided a common mechanism for pathway megalencephalies [43]. A National Institutes of Health (NIH) workshop proposed PIK3CA-related overgrowth spectrum (PROS) in 2013, with criteria published in 2015 [7, 44]. The somatic AKT1 p.(Glu17Lys) variant in Proteus syndrome was described by Lindhurst and colleagues in 2011 [5]. The relationship between PTEN, autism and extreme macrocephaly was established by Butler and colleagues in 2005, finding germline variants in 3 of 18 individuals with ASD and HC between +2.5 and +8.0 SD, precisely those with the largest HC [45].

1.3. Nomenclature and reclassifications

The designation M-CMTC was replaced by MCAP (or M-CM) because the lesions are reticular capillary malformations that fade and never ulcerate [40, 41]. MCAP belongs to PROS, which encompasses congenital segmental or focal overgrowth due to postzygotic activating PIK3CA variants, including dysplastic megalencephaly, hemimegalencephaly and focal cortical dysplasia; MPPH does not belong to PROS because it results from germline variants in other pathway genes [2, 7]. Smith's Recognizable Patterns of Human Malformation treats MCAP and MPPH as a single entity, whereas Mirzaa and GeneReviews keep them as related but distinct disorders, a separation that corresponds to different genes, mechanisms and recurrence risks [3, 40]. In PHTS, the 2026 ERN GENTURIS guideline recommends abandoning the terms Cowden, Bannayan-Riley-Ruvalcaba (BRRS), Lhermitte-Duclos, SOLAMEN and Proteus-like, and designating all carriers of a germline pathogenic PTEN variant as PHTS [10, 19].

1.4. Scope and cross-references

This article addresses the approach to the child with macrocephaly and the monogenic forms whose central manifestation is excessive brain growth. PHTS is addressed here only from the perspective of the macrocephalic child, without developing full diagnosis and cancer surveillance. ASD, Beckwith-Wiedemann syndrome, neurofibromatosis type 1, RASopathies and leukodystrophies form part of the differential diagnosis. Sotos, Weaver and Tatton-Brown–Rahman syndromes, PROS as a whole, Proteus, Simpson-Golabi-Behmel and Smith-Kingsmore syndromes, Canavan disease, Alexander disease and mucopolysaccharidosis type III are summarised here with regard to the differential diagnosis of macrocephaly.

2. Molecular Genetics and Pathophysiology

2.1. Two major molecular families

Monogenic macrocephaly with overgrowth groups into two families: epigenetic regulator genes and PI3K-AKT-mTOR pathway genes [5]. In PI3K signalling, cascade activation increases cell growth; activating variants are known in AKT1, AKT3, PIK3CA, PIK3R2 and MTOR, and inactivating variants in pathway inhibitors such as PTEN [5]. A third, less frequent category includes leukodystrophies and metabolic disorders with megalencephaly from substance accumulation or white matter oedema [2, 6]. The following table summarises the genetic bases of overgrowth according to OXF, supplemented by GeneReviews.

Sources: [3, 5, 11, 15].

2.2. The PI3K-AKT-mTOR pathway and the unifying cyclin D2 mechanism

Class I PI3K is a heterodimer formed by a catalytic subunit (p110α, encoded by PIK3CA) and a regulatory subunit (p85β, encoded by PIK3R2); the complex converts phosphatidylinositol-4,5-bisphosphate (PIP2) to phosphatidylinositol-3,4,5-trisphosphate (PIP3), recruiting and activating AKT kinases, which phosphorylate downstream targets including mTOR; cyclin D2 (CCND2), regulating the G1-S cell cycle transition, is a downstream pathway target [3]. PTEN is the phosphatase reversing PIP3 to PIP2, so its loss of function leaves phosphorylated AKT1 uninhibited [10]. In pathway megalencephalies, the mechanism is gain of function in PIK3CA, PIK3R2, AKT3, CCND2 and MTOR, and loss of function in PTEN [3, 7, 10]. MPPH-causing CCND2 variants cluster in the last exon around a GSK-3β-phosphorylated residue; mutant cyclin D2 resists proteasomal degradation and accumulates, with excessive neural progenitor proliferation; cells with PIK3CA, PIK3R2 or AKT3 variants show the same cyclin D2 accumulation, making this protein a convergence point for pathway megalencephalies [43, 46]. PIK3CA mediates insulin and insulin-like growth factor 1 (IGF1) action; constitutive activation may mimic these signals and explains hypoinsulinaemic hypoglycaemia when allele burden is high in target tissues [7]. In inherited metabolic disease classification, PIK3CA, PIK3R2 and PTEN are among phosphoinositide phosphorylation disorders [47].

Phenotype depends on the degree of pathway activation, variant timing and tissue distribution. Weakly activating, typically germline variants produce diffuse megalencephaly with ID and ASD; moderately activating variants in high-level mosaicism produce megalencephaly with pigmentary abnormalities; strongly activating variants in very low-level mosaicism produce hemimegalencephaly and focal cortical dysplasia, in a continuum from diffuse to segmental [2, 3]. Strongly oncogenic PIK3CA hotspot variants (codons 542, 545 and 1047) are associated with focal phenotypes (CLOVES, vascular malformations, hemimegalencephaly and focal cortical dysplasia) and, when present in the brain, more severe epilepsy; MCAP is predominantly associated with less activating variants distributed throughout the gene [7]. In a Dijon series of 162 tested PROS patients, strongly oncogenic variants were more frequent without brain overgrowth (50.6%) than with it (15.2%; p=0.00055) [48].

2.3. MPPH: PIK3R2, AKT3 and CCND2

Among the 41 molecularly confirmed MPPH individuals compiled by GeneReviews, PIK3R2 accounts for approximately 40%, AKT3 approximately 30% and CCND2 approximately 30% [3]. Most variants are germline de novo; somatic mosaicism has been described mainly in PIK3R2 and AKT3 [3]. PIK3R2 NM_005027.4:c.1117G>A p.(Gly373Arg) is the most frequent variant; it lies in the inter-SH2 domain and prevents p85β from restraining p110, increasing PIP3 and AKT-mTOR signalling [3, 24]. Mirzaa and colleagues screened 118 children with bilateral perisylvian polymicrogyria in 2015 and found PIK3R2 variants in 20 (19 p.Gly373Arg and one p.Lys376Glu), seven with isolated polymicrogyria without MPPH traits; PIK3R2 is thus the most frequently identified cause of isolated bilateral perisylvian polymicrogyria, and variant presence does not require the full phenotype [24]. In AKT3, the most frequent MPPH variant is NM_005465.7:c.1393C>T p.(Arg465Trp); half of pathogenic AKT3 alterations are gene-containing 1q43-q44 duplications or triplications detectable by chromosomal microarray, and somatic locus duplication has been found in hemimegalencephaly and focal cortical dysplasia [3]. AKT3 NM_005465.7:c.49G>A p.(Glu17Lys), paralogous to the Proteus syndrome AKT1 p.(Glu17Lys) variant, was found in mosaic form, only in brain tissue, in two patients with hemimegalencephaly [3, 49]. CCND2 variants are last-exon missense variants with a stabilising effect [43].

Allelic disorders confirm the pathway dosage effect: AKT3 loss of function (1q43q44 deletions) and proximal truncating CCND2 variants cause microcephaly with ID and, for CCND2, short stature, the phenotypic inverse of gain of function [3, 46]. ClinGen classifies the CCND2–MPPH relationship as definitive (27 May 2025) and assigns PIK3R2 haploinsufficiency and triplosensitivity scores of 0, consistent with a gain-of-function mechanism [24, 46]. Germline variant penetrance is estimated at 100%, with variable expressivity [3].

2.4. MCAP and focal megalencephalies: mosaic PIK3CA

In MCAP, PIK3CA variants are somatic in approximately 90% and germline in approximately 10% of patients; germline forms are associated with a milder phenotype, with megalencephaly and ID without vascular abnormalities or segmental overgrowth [2]. They are missense or in-frame deletion-insertions, never truncating, all gain of function; more than half of PROS variants are recurrent missense also known in oncology [7, 22]. Hemimegalencephaly, dysplastic megalencephaly and focal cortical dysplasia result from brain-restricted somatic PIK3CA, AKT3 or MTOR variants; dysplastic hemimegalencephaly corresponds to a large area of type II focal cortical dysplasia [2, 25]. In a Korean series, a somatic MTOR variant was found in brain tissue from 12 of 77 patients (15.6%) with type II focal cortical dysplasia [16].

ClinGen considers the association between germline PIK3CA variants and hereditary breast cancer refuted (21 December 2023); a germline PIK3CA finding does not indicate hereditary tumour predisposition [21]. For classifying somatic gain-of-function variants in AKT3, MTOR, PIK3CA and PIK3R2, the ClinGen Brain Malformations Expert Panel adapted American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria: 12 criteria do not apply (including PVS1, PM4 and PP3); PS2 was divided into PS2_moderate_1 (variant absent in parents with confirmed parentage) and PS2_moderate_2 (present in lesional tissue and absent or at a significantly lower allele fraction in another tissue), strong only with both; PM2 becomes supporting and applies only with ≤1 gnomAD individual; BS1 applies above 0.019% and BA1 above 0.093%; PM1 was downgraded to supporting [50]. Half of the 43 panel-classified variants were absent from ClinVar [50].

2.5. PTEN

PTEN is located at 10q23.31 (reference transcript NM_000314.8, in the MANE series, Matched Annotation from NCBI and EMBL-EBI). Variants are distributed throughout the gene (missense, nonsense, splicing, small indels and large deletions); approximately 40% lie in exon 5, encoding the catalytic phosphatase motif, and 76% predict truncated, absent or dysfunctional protein; many missense variants are functionally null and some have dominant-negative effects [10]. Missense variants are over-represented in PTEN-associated ASD, and developmental delay and macrocephaly are more associated with missense, whereas skin lesions, benign thyroid disease and cancer are more associated with truncating variants [10, 51]. Butler and colleagues' variants were described using one-letter protein nomenclature (H93R, D252G, F241S), corresponding to p.(His93Arg), p.(Asp252Gly) and p.(Phe241Ser) [45]. Penetrance is age-dependent and approaches 99% in adulthood, with no known non-penetrant carriers [10]. Contiguous PTEN and BMPR1A deletions cause juvenile polyposis of infancy, potentially mimicking BRRS [10].

2.6. MTOR and other pathway megalencephalies

Smith-Kingsmore syndrome results from constitutional or early postzygotic activating MTOR (NM_004958.4) variants; 20–25% of patients are mosaic, and c.5395G>A p.(Glu1799Lys) is present in almost 50% [14]. Approximately 86% of variants are de novo, 11% inherited from an unaffected parent with gonadal mosaicism and 3% from an affected parent [14]. Recessive megalencephalies include isolated megalencephaly from TBC1D7 loss of function (OMIM 248000; ORPHA:2477), HERC1-related megalencephaly–severe kyphoscoliosis–overgrowth syndrome (OMIM 617011; ORPHA:457359) and STRADA-related polyhydramnios–megalencephaly–symptomatic epilepsy syndrome (OMIM 611087; ORPHA:500533), the latter with early lethality [3, 9, 30, 52, 53].

2.7. Epigenetic regulators

NSD1 (5q35.3) encodes a histone methyltransferase for lysine 36 of histone H3 (H3K36); Sotos syndrome results from haploinsufficiency through intragenic variants or deletion [4, 32]. In Japan, the recurrent 1.9 Mb 5q35 microdeletion, mediated by non-allelic homologous recombination between low-copy repeats, is the most frequent finding; point variants predominate in other populations [4, 32]. Deletions are associated with more severe ID and more structural abnormalities, and an NSD1 methylation signature has high sensitivity and specificity [32]. NSD1 and SETD2 write the H3K36me mark and DNMT3A reads it, explaining phenotypic overlap between Sotos and SETD2- and DNMT3A-related Sotos-like forms [32].

EZH2 (7q36.1; NM_004456.4) is the catalytic PRC2 subunit, trimethylating lysine 27 of histone H3 (H3K27me3); Weaver variants are mainly missense, and truncating variants cluster in the last exon, sparing the SET domain [5, 11]. EED variants cause Cohen-Gibson syndrome and SUZ12 variants Imagawa-Matsumoto syndrome (OMIM 618786), both Weaver-like; one SUZ12 variant was inherited from a mosaic father [4, 37, 54]. DNMT3A (2p23.3) encodes a de novo DNA methyltransferase; TBRS results from loss of function through truncating variants, splice variants, whole-gene deletions or missense clustered in PWWP, ADD and methyltransferase domains [12]. NM_022552.5:c.2645G>A p.(Arg882His) (described in GeneReviews on NM_175629.2), an acute myeloid leukaemia (AML) somatic hotspot, also occurs in the TBRS germline, potentially with an additional dominant-negative effect [12]. NFIX (19p13.2) causes Malan syndrome through haploinsufficiency, with point variants clustered in exon 2 and 19p13.2 deletions in approximately 25% of cases; exon 6–10 variants escaping nonsense-mediated decay (NMD) cause allelic Marshall-Smith syndrome; in microdeletions, neighbouring CACNA1A, 109 kb away, contributes to seizures and episodic ataxia [13, 39]. CHD8 causes a neurodevelopmental disorder with overgrowth through heterozygous loss of function [55]. In the British overgrowth-with-ID cohort (n=710), six epigenetic genes (NSD1, EZH2, DNMT3A, CHD8, HIST1H1E, EED) explained 44% of cases [56].

2.8. Megalencephalic leukodystrophies

In MLC, MLC1 accounts for approximately 76% of cases; the mechanism is astrocytic, with impaired volume and channel regulation in perivascular endfeet and myelin vacuolation [29]. Biallelic MLC1 or HEPACAM (GLIALCAM) and heterozygous GPRC5B variants cause classical disease; heterozygous gain-of-function HEPACAM or biallelic AQP4 variants cause the improving form [8, 57]. Alexander disease results from heterozygous, usually de novo, GFAP variants [1, 58]. Canavan disease is AR, due to aspartoacylase deficiency (ASPA) with N-acetylaspartate (NAA) accumulation, myelin disruption and spongiform white matter degeneration [1].

2.9. Reference variants

Descriptions follow Human Genome Variation Society (HGVS) nomenclature. Variants with genomic coordinates were normalised with VariantValidator (GRCh38, MANE Select); the others follow the GeneReviews descriptions, which state that variants were provided by the authors [3, 7, 59]. The MLC1 founder variants, normalised on the MANE transcript NM_015166.4, are c.135dup p.(Cys46LeufsTer34) in the Agrawal community of India, c.176G>A p.(Gly59Glu) in Libyan and Turkish Jews, c.278C>T p.(Ser93Leu) in Japanese people, c.824C>A p.(Ala275Asp) in Koreans and c.908_918delinsGCA p.(Val303GlyfsTer96) in Egyptians [29]. The Ashkenazi ASPA variant known by the historical designation E285A corresponds to NM_000049.4:c.854A>C p.(Glu285Ala) and accounts for 80–85% of alleles in that population [58].

3. Clinical Spectrum and Manifestations by System

3.1. Normal head circumference growth

HC increases approximately 1 mm daily between 26 and 32 weeks' gestation and approximately 0.7 mm daily between 32 and 40 weeks; term newborn HC is 32–37 cm and brain weight approximately 370 g [2]. During the first year, HC grows approximately 2 cm monthly in the first quarter, 1 cm monthly in the second and 0.5 cm monthly in the second half (≈12 cm total); by the end of early childhood it reaches approximately 90% of adult size, 52–58 cm in women and 52.5–58.5 cm in men, with a brain weighing approximately 1500 g [2]. These values allow recognition of acceleration suggesting hydrocephalus and progressive megalencephaly in PI3K pathway disorders, whose excessive growth is concentrated in the first two years [2, 3, 7].

3.2. Benign familial macrocephaly and benign enlargement of the subarachnoid spaces

Benign familial macrocephaly is diagnosed by increased parental HC and absence of other findings and developmental delay; inheritance may be AD [1]. Isolated benign megalencephaly is familial in more than half of cases according to Accogli and colleagues' review, citing earlier studies [2]. Benign enlargement of the subarachnoid spaces (BESS, also termed benign external hydrocephalus) is the most frequent cause of infantile macrocephaly, with an estimated incidence of 0.5 per 1000 live births, male predominance and familial recurrence of benign macrocephaly in approximately 40% (data cited by Accogli and colleagues) [2]. HC accelerates around 6 months, stabilises near 18 months and enlargement resolves spontaneously by 3 years; a craniocortical width above 10 mm suggests pathology, and the cortical vein sign distinguishes BESS from a subdural collection [2]. The most serious complication is subdural haematoma from stretching bridging veins; treatment is conservative, and shunting or temporary acetazolamide have been used in isolated cases [2]. Virchow-Robin spaces (perivascular extensions of the subarachnoid space) are normal, common magnetic resonance imaging (MRI) findings, sometimes prominent and generalised in children with macrocephaly [1]. OXF cautions that PTEN and PTCH1 variants may present with few syndromic signs, so labelling benign familial macrocephaly requires careful proband and parental examination [1].

3.3. MCAP

Head and brain growth. Most patients have macrocephaly from megalencephaly at birth; some are born with normal HC and develop progressive megalencephaly in the first year [7]. In a review of 21 children with MCAP, birth HC ranged from +2 to +7 SD; SD increased during the first year in most and, although it may stabilise in early childhood, typically remains ≥+3 SD; adult HC ranges from +2 to +10 SD [7]. After ventricular shunting, head growth remains accelerated, proving that megalencephaly is primary rather than secondary to hydrocephalus [7, 40]. More than 90% have universally progressive congenital megalencephaly, sometimes detected on prenatal ultrasound with ventriculomegaly or a thick corpus callosum [7]. In the French series of 33 patients with PIK3CA mosaicism (Garde 2021; median age 9.9 years), macrocephaly was present at birth in 20 of 31 and at follow-up in 31 of 32; mean HC increased from +2.6 SD at birth to +4.7 SD at last follow-up, concentrated in the first two years [22, 63].

Neuroimaging. In a series of 65 children with MCAP (Conway 2007), 37 (56%) had ventriculomegaly ranging from mild to overt hydrocephalus, and 15 had tonsillar ectopia with or without herniation and syringomyelia; more than half underwent shunting or third ventriculostomy, generally in the first year [7]. Unlike ventriculomegaly, tonsillar ectopia is rarely congenital; the cerebellum may grow faster than the cerebrum during the first two years, explaining acquired ectopia in a crowded posterior fossa [2, 7]. Polymicrogyria is present in more than half, mainly bilateral perisylvian but also bilateral frontal or focal, and increases risks of epilepsy, oromotor dysfunction, delay and tone abnormalities [7]. Frequencies vary by series and imaging review: ventriculomegaly 47–71%, cerebral asymmetry 23–52%, abnormal corpus callosum 5–71%, polymicrogyria 9–71%, Chiari 23–86% and abnormal white matter 24–65% [22]. In the French series, among 21 reviewed MRI scans, 20 showed megalencephaly, 16 a thick corpus callosum, 12 Chiari, 10 ventriculomegaly or hydrocephalus, 6 cerebral asymmetry and 2 polymicrogyria [63]. In a Korean cohort of 12 patients with confirmed variants (Seoul National University Hospital, 1998–2019), 11 of 11 with MRI had megalencephaly or hemimegalencephaly, 10 of 11 (91%) Chiari type 1, 2 of 11 polymicrogyria, and 3 of 12 (25%) underwent ventriculoperitoneal shunting [60]. OXF and Smith also describe cavum septum pellucidum or cavum Vergae, dilated dural venous sinuses, prominent Virchow-Robin spaces and lumbar syrinx [40].

Neurodevelopment and epilepsy. Most MCAP patients have some ID and mild to moderate delay with continuous progress; fewer than 10% have severe disability [7]. In the 2019 French series summarised in the national protocol (PNDS), motor delay occurred in 60%, language delay in 65%, ID in 52%, behavioural disorders in 37% and epilepsy in 18% [22]. Among Mirzaa's 21 children in 2012, six had autistic traits or an autism diagnosis [7]. GeneReviews estimates epilepsy in 30–40% of individuals with PIK3CA variants, depending on tissue distribution and cortical malformations, and the SESAM protocol reports seizures in only 10–15% of MCAP patients in the French series; the figures reflect different populations and definitions [7, 64]. In the Korean series, 5 of 12 had delay or ID and 2 of 12 seizures [60]. Motor dysfunction is multifactorial (megalencephaly, malformations, hypotonia, asymmetry, connective tissue dysplasia); young children with frontal polymicrogyria are hypotonic with pseudobulbar signs, and older children develop spasticity [7]. Tonsillar ectopia manifests in infants as irritability, drooling, dysphagia and respiratory problems, particularly central apnoea, and in children as neck pain or headache, weakness, sensory, visual, swallowing and behavioural changes [7].

Skin, connective tissue and limbs. Capillary malformations (particularly midfacial naevus flammeus on the nose and philtrum, and a reticular pattern on the body) are present at birth and may fade until unnoticed [7, 40]. In the French series, cutaneous vascular malformations occurred in 100% (capillary malformation 85%, cutis marmorata 52%), body or facial hemihypertrophy in 64% (leg length discrepancy 54%), soft skin in 56%, skin hyperelasticity in 43%, joint hypermobility in 55%, 2–3 toe syndactyly in 42%, polydactyly in 21% and a wide gap between first and second toes in 31% [22]. Facial features include high forehead, frontal bossing, dolichocephaly, bitemporal narrowing, hypertelorism, flat nasal bridge and asymmetry [22].

Endocrinology, heart and tumours. Hypoinsulinaemic hypoketotic hypoglycaemia, with undetectable insulin during episodes, mimics congenital hyperinsulinism; it is generally neonatal, but one boy with MCAP had his first episode at 6 years [7]. Hypothyroidism, growth hormone (GH) deficiency, central adrenal insufficiency and type 1 diabetes, as well as structural heart defects (atrial and ventricular septal defects) and great vessel abnormalities, are also described [7, 22]. The most common benign tumours are vascular, including cerebral cavernous haemangiomas; two MCAP patients (21 months and 5 years) had meningioma [7]. Wilms tumour has an estimated PROS frequency of 1.4–3.3%; of 12 individuals with Wilms tumour or nephroblastomatosis compiled by GeneReviews, 8 had CLOVES, 2 MCAP and 2 Klippel-Trenaunay [7]. Leukaemia, vestibular schwannoma and retinoblastoma have been reported individually without data allowing an association; one MCAP patient with p.(His1047Tyr) developed diffuse large B-cell lymphoma in late childhood [7, 61]. No neoplasm developed during follow-up with abdominal ultrasound surveillance in the Korean series [60].

3.4. MPPH

MPPH is characterised by megalencephaly and bilateral perisylvian polymicrogyria; HC at birth ranges from normal to +6 SD and, at later ages, from +3 to +10 SD [3]. Megalencephaly is congenital or early postnatal, rapidly progressive in the first year, and brain overgrowth persists after shunting [3]. Almost all patients have ventriculomegaly and about 50% overt hydrocephalus requiring shunting; with limited retrospective data, the risk of hydrocephalus and of tonsillar ectopia with compression of the lower brainstem and upper spinal cord is highest in the first two years [3]. Polymicrogyria is present in all, almost always bilateral perisylvian; a thick corpus callosum and tonsillar ectopia or Chiari malformation may coexist [3]. Oromotor dysfunction (expressive language delay, profuse drooling, dysphagia) affects most patients and sometimes leads to gastrostomy; hypotonia is present at birth in most and may remain severe; ID is almost universal, mild to severe, and its severity depends on the extent of the cortical malformations and on the age at onset and severity of epilepsy [3]. The frequency of epilepsy is inconsistent within GeneReviews itself: the summary and the neurological description report early-onset epilepsy in about 50% (focal to generalised, sometimes infantile spasms and refractory), whereas Table 2 gives about 100% for epilepsy and about 50% for oromotor dysfunction, which the text describes in most; the values adopted are those of the text (epilepsy ~50%, oromotor dysfunction in most) [3]. Postaxial polydactyly, of one to four extremities, occurs in 50%; visual problems (cortical visual impairment, blindness) are common; heart defects, endocrinopathies (hypoglycaemia, GH deficiency, hypothyroidism, Hashimoto thyroiditis) and kidney abnormalities have each been described in fewer than five individuals [3].

Medulloblastoma has been described in three individuals with MPPH or a CCND2 variant, suggesting a causal link [3, 46]. Gene-specific correlations exist: CCND2 variants are associated with more extensive polymicrogyria (frontal and occipital lobes), more severe epilepsy and ID and more frequent polydactyly; AKT3 is associated with connective tissue laxity and capillary malformations overlapping MCAP; PIK3R2 frequently produces ventriculomegaly and milder cognitive impairment, and somatic PIK3R2 forms may cause polymicrogyria with normal HC; somatic AKT3 may cause megalencephaly without polymicrogyria [2, 3]. OXF includes macrocephaly among polymicrogyria categories as a marker of PIK3CA pathway disorders [65]. MPPH differs from MCAP through absence of vascular malformations, syndactyly, segmental overgrowth, cerebral asymmetry and connective tissue dysplasia [22, 41].

3.5. Hemimegalencephaly and dysplastic megalencephaly

Hemimegalencephaly is overgrowth of one hemisphere, with or without cortical dysplasia, accompanied by cognitive deficits, frequent seizures, focal deficits and facial asymmetry; dysplastic megalencephaly is focal overgrowth with cortical dysplasia, sometimes bilateral, with delay and severe epilepsy in the first months (focal seizures, infantile spasms, tonic seizures or an Ohtahara phenotype) [7]. Focal cortical dysplasia types I, II and III show dysmorphic neurons, balloon cells and laminar disorganisation, with less overgrowth and sometimes refractory epilepsy [7]. Imaging distinguishes classical hemimegalencephaly, hemi-hemimegalencephaly or quadrantic dysplasia (up to three lobes, with the “occipital sign”) and total hemimegalencephaly including the brainstem and cerebellum [2]. Epilepsy is almost universal in hemimegalencephaly [22].

3.6. PHTS in children

Macrocephaly from megalencephaly is almost universal and frequently severe: HC often exceeds +5 SD and is more pronounced in PTEN-related than idiopathic ASD [10]. It is present in 81–100% of children and 61–79% of adults, a difference attributed to recognition of mild adult phenotypes; approximately two thirds of children have HC between +4 and +6 SD, and mean adult HC is 60.0 cm in women and 62.8 cm in men, with normal height and weight [2, 66]. In BRRS, Smith records mean HC +4.89 SD with normal-sized ventricles, and OXF reports HC frequently above +4.5 SD and birth weight generally above 4000 g [1, 67]. Approximately half of PHTS children have a neurodevelopmental disorder; in a large study, frequency was 35% in children and 23% in adults, with ASD generally mild to moderate; the cognitive phenotype includes language delay, poor working memory and processing speed, and adaptive and sensory abnormalities [10]. Psychiatric disorders affect 11–44% (anxiety, depression, bipolar disorder, attention deficit hyperactivity disorder, obsessive-compulsive disorder, psychosis) [10]. Epilepsy occurs in 6–17%, versus approximately 1% in the population, associated with focal cortical dysplasia, polymicrogyria and cavernomas and more frequent with ASD [10, 66]. In BRRS, hypotonia, gross motor delay with mild proximal myopathy in 60%, speech delay in 50%, ID in 15–20% and seizures in 25% are described [67].

On the skin, café-au-lait macules and penile freckling appear in childhood, and lipomas and linear naevi accompany overgrowth; penile macules are not always present at birth and become more evident in late childhood [10, 67]. Thyroid nodules are common and carcinoma, particularly follicular, has been described from 4 years (GeneReviews) or 7 years (according to the American Association for Cancer Research, AACR); gastrointestinal polyps affect 25–30% in a paediatric series, from mid-childhood to adolescence [10, 51]. In a retrospective paediatric study, 66% had dermatological, 34% gastrointestinal and 26% abnormal thyroid imaging findings [2]. Neuroimaging is abnormal in at least half (symmetrical megalencephaly, thick corpus callosum, tonsillar ectopia, enlarged perivascular spaces, white matter abnormalities, developmental venous anomalies) [2]. Lhermitte-Duclos disease (dysplastic cerebellar gangliocytoma, with a “tigroid” pattern) occurs in 5–10% (11% of women and 4% of men), generally in the third and fourth decades [2, 66]. Tonsillar problems occur in 58% of children, 74% of whom undergo surgery [66].

3.7. Smith-Kingsmore

In more than 100 individuals, macrocephaly (HC ≥2 SD) is present in approximately 90% and is the distinguishing trait; delay or ID, generally moderate to severe, affects approximately 95%, with regression in approximately 18%, and one third never acquire walking [14, 16]. Epilepsy (generalised, focal or status epilepticus) affects approximately 61%, more severe with malformation; brain malformations (megalencephaly, polymicrogyria, focal cortical dysplasia, hemimegalencephaly, corpus callosum abnormalities, ventriculomegaly) affect approximately 79% [14, 16]. Hypotonia (~79%), sleep disturbances (~77%), ASD (~46%), hyperphagia (~41%) and stable aortic root dilatation are also common; hyperpigmented naevi and pigmentary mosaicism indicate mosaic disease [16]. Adult survival is common, with one patient alive at 70 years [14].

3.8. Epigenetic overgrowth syndromes

Sotos. The cardinal triad, present in ≥90% of more than 266 individuals with an NSD1 variant, comprises facial appearance (broad forehead, dolichocephaly, sparse frontotemporal hair, downslanting palpebral fissures, malar flushing, prominent chin), learning difficulty and overgrowth (height and/or HC ≥2 SD) [4]. Macrocephaly is congenital in most, with increased head growth velocity in the first year, and persists in adulthood; Smith quantifies it at 50% at birth, 100% at one year and 86% in adults [4, 32]. Height may normalise in adulthood (mean 182 cm in men and 174 cm in women), and bone age is advanced in 75–80% of prepubertal children [4, 32]. MRI is abnormal in most, particularly trigone-predominant ventricular dilatation, midline abnormalities (corpus callosum hypoplasia or agenesis, mega cisterna magna, cavum septum) and enlarged extracerebral spaces [4, 32]. Intelligence quotient ranges from 40 to 129 (mean 78), verbal exceeding performance; heart defects occur in 15–40%, kidney abnormalities in approximately 15%, as well as seizures (half febrile), scoliosis, hyperinsulinaemic hypoglycaemia and maternal pre-eclampsia in approximately 15% of pregnancies [4, 32]. Neoplasia risk is approximately 2.2% according to Smith and approximately 3% according to Accogli's review [2, 32].

Weaver and PRC2 overgrowth. Growth and bone maturation accelerate prenatally; tall stature is almost uniform, and final height reaches 194.2 cm in men and 176.3 cm in women (>+3.5 SD), with adult HC of 61 cm and 59.5 cm, respectively [11, 54]. Macrocephaly is less consistent than in Sotos: Smith reports 83%, but among 59 individuals with GeneReviews data, 32 had HC >2 SD and 27 <2 SD, with values up to +5.5 SD and birth HC ≥2 SD in 5 of 6 infants; HC is large but proportional to height [11, 54]. Intellect ranges from normal to severe ID, and the gestalt (round face, hypertelorism, “stuck-on” chin with a horizontal crease) is most recognisable between 1 and 3 years, with camptodactyly, doughy skin, a hoarse low-pitched cry and umbilical hernia [4, 11]. Ventriculomegaly has been described in seven individuals, generally with normal CSF pressure and no shunt requirement [11]. Neuroblastoma frequency is increased (5–7% in the GeneReviews table), without evidence of increased other neoplasms [11, 54].

Tatton-Brown–Rahman. Overgrowth is present in more than 80%, tall stature in approximately 70% (−0.3 to +6 SD), macrocephaly in approximately 50% (HC −1.2 to +8.0 SD; mean +2.3 SD at birth) and overweight or obesity in approximately 65% [12]. ID was mild in 28%, moderate in 58% and severe in 14% of 83 individuals; formally assessed autism was confirmed in 8 of 18 (44%); kyphoscoliosis affects approximately 30% and seizures approximately 20% [12]. The round coarse face, with thick low horizontal eyebrows, narrow palpebral fissures and prominent upper central incisors, becomes more recognisable in adolescence [12]. Eighteen studied individuals showed macrocytosis without anaemia, relative lymphocyte reduction and neutrophil increase, with an altered haematopoietic methylome [68]. Among eight TBRS patients developing neoplasia, five had variants at Arg882 [12].

Malan. Macrocephaly was described in all patients, with HC >2 SD in approximately 41% of newborns and ≥77% of adults; height exceeds +2 SD in 56% [13]. ID is typically moderate to severe, with an anxious profile in more than 80%, frequent ophthalmological abnormalities (blue sclerae ~70%, refractive errors ~75%) and a marfanoid habitus; MRI shows ventricular dilatation and corpus callosum hypoplasia in 8 of 16 and Chiari type 1 in 6 of 16 [13, 39].

CHD8. Macrocephaly, more frequent in the first year, and tall stature, particularly at puberty, are the commonest findings, with ASD and/or delay, sleep disturbances, constipation, hypotonia in approximately 30% and seizures in 10–15% [15].

3.9. Megalencephalic leukodystrophies

In classical MLC, macrocephaly sometimes appears at birth and more frequently in the first year, reaches +4 to +6 SD and then grows in parallel, several SD above the 98th percentile; it is the presenting sign in approximately 95% [8, 29]. Early development is normal in most, followed by slowly progressive ataxia, spasticity, sometimes extrapyramidal signs, and late, generally mild cognitive decline; independent walking is generally lost in the first or second decade, with wide variation [8, 29]. Approximately 75% have at least one seizure before 20 years, approximately 60% epilepsy and 15–20% status epilepticus; minor head trauma may precipitate seizures, status epilepticus, prolonged unconsciousness or acute motor deterioration [29]. MRI shows abnormal swollen hemispheric white matter and anterior temporal subcortical cysts, frequently also frontoparietal, with relatively mild physical signs [8, 58]. In the improving form, presentation is similar but followed by improvement; macrocephaly generally persists, and some retain stable ID or ASD [8]. Infantile Alexander disease, beginning within the first two years, combines megalencephaly (sometimes hydrocephalus), psychomotor regression, seizures and spasticity, with frontal-predominant white matter signal abnormalities and death in the first decade [1]. Canavan disease manifests in the first year with hypotonia progressing to spasticity, macrocephaly, lack of head control and progressive cognitive delay [58].

4. Clinical Recognition and Warning Signs

4.1. Why measuring and interpreting head circumference matters

The American Academy of Pediatrics (AAP) recommends HC measurement at surveillance visits up to 2 years, and measurement should be repeated at any age with signs of intracranial hypertension (ICH), delay or neurological deficit [2]. The international PHTS consensus recommends HC measurement at least annually up to 5 years in children with a PTEN variant: deceleration should prompt growth restriction investigation, and acceleration urgent imaging and neurosurgical assessment to exclude hydrocephalus [66]. Macrocephaly is associated with neurodevelopmental abnormalities: in an Israeli community study of more than 4000 children with neurodevelopmental disability, 1.4% had HC >98th percentile, and macrocephaly plus disability was associated with greater seizure risk (odds ratio [OR] 7.7); idiopathic infantile macrocephaly was associated with greater ASD risk (OR 5.44) in a British study; up to 15% of people with ASD have macrocephaly (data cited by OXF and Accogli) [1, 2].

4.2. Emergency signs: intracranial hypertension

The first step in a macrocephalic child is to exclude ICH, a neurosurgical emergency [2]. Warning signs are progressive HC increase or rapid percentile crossing, tense or bulging fontanelle, extreme irritability or drowsiness, ocular changes (strabismus, ptosis, blurred or double vision, anisocoria, downward “setting-sun” gaze), nausea and vomiting, gait change, loss of consciousness and seizure [2]. A fontanelle bruit or high-output heart failure suggests an intracranial vascular malformation such as vein of Galen malformation [2]. In PHTS, positional or projectile vomiting, abnormal neurological examination, new signs, ICH signs or new-onset headaches in a child under 5 years warrant gadolinium MRI and MR angiography or venography [10]. In PI3K pathway megalencephalies, central apnoea, dysphagia, drooling and irritability in infants, and neck pain, headache, weakness or sensory changes in children suggest progressive tonsillar ectopia; worsening seizures may reflect ICH or hypoglycaemia [7, 22].

4.3. When to suspect a genetic cause, by age

Prenatal period. MCAP megalencephaly is mainly detected in the third trimester because it may appear only after 32 weeks, and polymicrogyria is rarely seen on ultrasound; fetal MRI is useful after 30 weeks [22]. In a Chinese prospective cohort of 87 fetuses with HC ≥+2 SD and normal microarray, trio exome sequencing established diagnosis in 34 (39.1%): 26 of 40 (65.0%) non-isolated macrocephalies and 8 of 47 (17.0%) isolated cases (p<0.001), predominantly mTOR pathway disorders, RASopathies and Sotos; yield was higher with diagnosis before 32 weeks, HC ≥+3 SD, micromelia and absence of a large-for-gestational-age fetus [69]. Maternal pre-eclampsia occurs in approximately 15% of pregnancies with Sotos fetuses [4].

Newborns and infants. Warning signs are birth HC ≥+3 SD, rapidly progressive first-year megalencephaly, midfacial and philtral capillary malformations, 2–3 toe syndactyly, postaxial polydactyly, body asymmetry, hypotonia, hypoglycaemia and oromotor dysfunction [3, 7, 22]. Neonatal jaundice, hypotonia and feeding difficulties, with birth length near the 98th percentile, suggest Sotos [4]. Penile macules, lipomas, vascular malformations and birth weight above 4000 g suggest PHTS [1, 67]. First-year macrocephaly followed by unsteady gait and falls suggests MLC [8, 29]. Congenital macrocephaly may be the initial isolated marker of glutaric aciduria type 1 and other cerebral organic acidurias, and cephalohaematomas occur in glutaric aciduria type 1 and Menkes disease [6].

Children and adolescents. In a child with ASD or delay and macrocephaly, particularly extreme, PHTS should be sought; Smith emphasises that severe progressive macrocephaly appears a good predictor of a PTEN variant in this subgroup and penile freckling should always be sought [45, 67]. Developmental regression, coarse facial features, organomegaly and white matter abnormalities point towards a metabolic disorder or leukodystrophy [1, 2]. Tall stature with advanced bone age and learning difficulty suggests an epigenetic overgrowth syndrome, and Sotos diagnosis is easier in early childhood, so old photographs should be requested when assessing adults [5].

4.4. Newborn screening

The sources consulted describe no newborn screening for the monogenic forms of macrocephaly addressed in this article; diagnosis depends on serial HC measurement and recognition of warning signs. Glutaric aciduria type 1 is a metabolic differential in which macrocephaly may be the first sign [6].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Measurement and reference charts

HC is measured with a non-stretch tape along the largest diameter, from the glabella or mid-forehead to the occipital prominence, and plotted on age- and sex-specific charts; at birth it should be standardised for gestational age [2]. World Health Organization (WHO) charts extend to 5 years and Centers for Disease Control and Prevention (CDC) charts to 36 months, with HC data up to 20 years in the 2000/2010 versions [2]. Values should be expressed in SD (z-score), not percentile alone, because +3, +5 or +10 SD limits described in megalencephalies cannot be distinguished on percentile charts [2, 70]. Interpretation compares HC with height and weight (isolated, relative macrocephaly or part of generalised overgrowth) and growth velocity (an abrupt increase suggests hydrocephalus) [2]. Parental HC measurement and plotting relative to height using adult centiles are mandatory [1]. Causes of falsely large head appearance, such as hairstyle, craniosynostosis and other skull shape changes, should be excluded [1].

5.2. Targeted history and examination

History includes a three-generation pedigree specifically asking about large HC in the family and neoplasms (breast, colonic polyps, endometrium and thyroid, suggesting PHTS; early or multiple basal cell carcinomas and medulloblastoma, suggesting Gorlin syndrome), consanguinity, birth HC, developmental milestones, regression and seizures [1]. In overgrowth, parental birth weights and growth, gestational diabetes, assisted reproduction, polyhydramnios, birth parameters, neonatal hypoglycaemia and accelerated growth periods are added [5]. Examination seeks coarse facial features (storage disorders), prominent forehead with a long face (Sotos), café-au-lait macules, haemangiomas and depigmented lesions, lipomas, vascular malformations and trichilemmomas (PHTS), perinasal and perioral papillomas (Cowden, Costello), limb asymmetry and syndactyly, philtral capillary malformation (MCAP), organomegaly, pigmented penile or vulval macules, ear creases and pits and macroglossia (Beckwith-Wiedemann), midline tongue groove and supernumerary nipples (Simpson-Golabi-Behmel), plantar connective tissue naevi (Proteus), scoliosis and umbilical hernia [1, 2, 5].

5.3. Neuroimaging

Neither the AAP nor the American College of Radiology has recommendations for imaging in macrocephaly [2]. Sampson and colleagues' algorithm, cited by Accogli, recommends imaging for developmental delay, neurological signs or symptoms, unexplained irritability, feeding changes, suspected abusive trauma or crossing two major percentiles; suggests considering it with dysmorphism, overweight or cutaneous or vascular signs; and allows deferring MRI with surveillance in the absence of risk factors and with a family history of macrocephaly [2]. Transfontanelle ultrasound is the initial approach while the fontanelle is open; MRI is the most accurate modality and may be performed in infants without sedation after feeding and wrapping; computed tomography (CT) is reserved for emergencies or bone structures, and rapid MRI or very low-dose CT serves hydrocephalus follow-up [2]. OXF considers neuroimaging particularly important in a child with macrocephaly and delay, preferring MRI for white matter, and indicates it in ID whenever microcephaly, macrocephaly or neurological signs are present [1, 71].

MRI interpretation follows two steps [2]. The first excludes secondary causes: hydrocephalus, enlarged extra-axial spaces, masses or cysts and bone dysplasia. The second distinguishes developmental from metabolic megalencephaly: only developmental disease is unilateral or asymmetrical and accompanied by cortical developmental malformations, whereas metabolic abnormalities are dynamic and symmetrical, with restricted diffusion or cysts. Developmental megalencephaly is subclassified as bilateral symmetrical or asymmetrical, hemimegalencephaly, hemi-hemimegalencephaly and total hemimegalencephaly [2]. Ventricular volume should be assessed relative to the degree of megalencephaly, because large ventricles are common in macrocephalic children [1]. Proton MR spectroscopy is diagnostic in few disorders, including Canavan disease with a high NAA peak [6]. In MCAP, the French PNDS recommends systematic initial brain MRI with a submillimetre 3D T1 sequence, T2 in three planes, axial fluid-attenuated inversion recovery (FLAIR) after 2 years and, when possible, contrast 3D T1, susceptibility sequences (SWI/T2*) and arterial spin labelling (ASL) perfusion [22]. In MPPH, GeneReviews recommends baseline MRI with diffusion to detect early neoplastic transformation in dysplastic cerebellar tissue and contrast if medulloblastoma is suspected, and whole-spine MRI with hydrocephalus or tonsillar ectopia [3]. In PHTS, routine brain MRI is not recommended without signs or symptoms [10, 66].

5.4. Diagnostic algorithm

Accogli and colleagues' algorithm (review Figure 14, viewed at https://pmc.ncbi.nlm.nih.gov/articles/PMC8795981/figure/F14/) starts from history, examination and neuroradiologist-reviewed MRI and divides investigation according to dominant findings [2].

Table 1 of the same review indicates blood testing for somatic MTOR, AKT3 and PIK3CA variants in hemimegalencephaly, inconsistent with its own text noting that somatic variants are not detected in blood DNA; in hemimegalencephaly, resected brain tissue has the highest detection rate [2, 7]. OXF proposes a similar sequence: neuroimaging, developmental assessment, metabolic screening (urinary organic acids and mucopolysaccharides), fragile X, genomic analysis (array, panel, exome or genome) with suggestive malformations, very deep sequencing (>1000×) of affected tissue DNA with strong suspicion of PIK3CA/AKT pathway disease, and PTEN analysis when PHTS is clinically compatible [1].

5.5. Genetic test selection

The American College of Medical Genetics and Genomics (ACMG, 2021) and AAP (2025) recommend exome or genome sequencing as first- or second-line testing in children with delay, ID and/or multiple congenital abnormalities [10, 16]. Accogli and colleagues prefer exome sequencing, initially analysing a virtual macrocephaly panel then expanding, over a standalone panel [2]. Chromosomal microarray is appropriate for sizing AKT3-containing 1q43-q44 duplications and chromosomal overgrowth abnormalities (4p duplication including FGFR3, 11p15 duplication, Pallister-Killian 12p tetrasomy, 15qter duplication or triplication including IGF1R, 5q35, 19p13.2 and 22q13 deletions) [3, 5]. In the British cohort of 710 individuals with overgrowth and ID, a cause in one of 14 genes was found in 50% (353/710), and exome sequencing had high yield [56].

PTEN. No clinical criteria establish PHTS diagnosis; molecular testing is necessary, and a Cleveland Clinic risk calculator exists [2, 10]. The 2023 American Association for Cancer Research (AACR) paediatric workshop recommends PTEN testing in a child with macrocephaly (>2 SD) and at least one of: ASD or developmental delay; PHTS dermatological findings; vascular malformation; gastrointestinal polyps; testing should also be considered for paediatric multinodular goitre, differentiated thyroid carcinoma or germ cell tumour with macrocephaly [51]. The sequential strategy is PTEN sequencing, then if negative deletion/duplication analysis and subsequently promoter sequencing; coding region sequencing detects up to 80% of variants, dosage analysis 3–11% and promoter analysis approximately 10% [10]. Smith emphasises that at least 10% of BRRS variants are intragenic or large deletions, requiring sequencing and dosage analysis [67]. Only approximately 25% of individuals with a PHTS-like phenotype have a germline P/LP PTEN variant; some have AKT1, PIK3CA or WWP1 variants, and somatic mosaicism should be considered when clinical criteria are strong and germline testing negative [51, 66]. A PTEN variant of uncertain significance (VUS) neither establishes nor excludes diagnosis [10]. The historical proportion of BRRS patients with a PTEN variant differs between sources (60% in OXF, ~70% or more in Smith), reflecting cohorts and detection methods from different periods [1, 67].

MPPH. Clinical diagnosis rests on two core findings, megalencephaly and polymicrogyria, and molecular diagnosis on a heterozygous pathogenic AKT3, CCND2 or PIK3R2 variant [3]. The multigene panel should include these and differential diagnosis genes and be optimised for mosaicism, because many commercial panels analyse blood only with a technical limit for low allele fractions; if exome or genome sequencing is negative, mosaicism-sensitive sequencing of saliva or affected or unaffected skin should be performed [3]. Sequencing detects 50% of AKT3 alterations (the other 50% are microarray-detectable duplications) and 100% of CCND2 and PIK3R2 alterations; the proportion of variants detectable by a method is not clinical sensitivity, and non-detection does not exclude clinical diagnosis [3].

MCAP and PROS. PROS diagnosis requires suggestive findings and a heterozygous activating mosaic or, rarely, constitutional PIK3CA variant [7]. DNA should come from clinically affected tissue (fresh skin biopsy over an affected area, surgical specimen, uncultured fibroblasts); blood is not recommended because PIK3CA was identified in blood in only 2 of 24 MCAP individuals, both with an apparently germline de novo variant [7]. In MCAP, saliva and skin fibroblasts from affected or unaffected skin have higher detection rates than blood; in focal cerebral forms, brain tissue obtained at epilepsy surgery has the highest rate [7]. Standard-depth exome sequencing may miss mosaic variants, particularly in blood; Sanger sequencing detects allele fractions only around 20%; targeted capture of the whole coding region with very deep next-generation sequencing (NGS) is preferable, and droplet digital polymerase chain reaction (PCR) or customised assays may be necessary; PCR assays designed for oncological hotspots miss rare MCAP variants [7]. Deletion/duplication analysis is not recommended because all variants are gain of function [7].

The Dijon laboratory used ultradeep NGS across several tissues in 162 referred patients and found the causal variant in 66.7% (108/162), with allele fractions as low as 1%; yield was 74% in syndromic and 35.5% in isolated cases (p=9.03×10⁻⁵), and allele fraction was higher in skin and overgrown tissues than blood and buccal mucosa (p=3.9×10⁻²⁵), regardless of phenotype [48]. This yield measures detection at a reference laboratory in a referred population, not clinical test sensitivity. In the Korean series, all patients tested in skin or buccal mucosa were diagnosed, but blood was positive in only 2 of 5; two had negative blood and positive skin Sanger testing, one exome missed a variant found by a deep panel, and one blood trio exome was negative with positive skin; mean allele fraction was 22.2% (6.3–35.3%) and did not correlate with severity; before diagnosis, 8 of 12 had undergone a mean 2.8 tests, five for Beckwith-Wiedemann [60]. CSF cell-free DNA may be an alternative without accessible lesional tissue: in one patient with p.(His1047Tyr), allele fraction was 3.08% in CSF, 37.31% in fibroblasts and 2.04% in blood [61]. A negative result in one sample does not exclude diagnosis and may warrant another tissue sample [7, 22].

Smith-Kingsmore, epigenetic syndromes and leukodystrophies. In Smith-Kingsmore, 20–25% of patients are mosaic, so a fresh skin biopsy may need testing [14]. In Sotos, NSD1 is found in only 60–90% of patients with a clinical Sotos or Sotos-like diagnosis, depending on the clinical definition used; targeted dosage analysis detects an additional 1–5% of partial deletions, and the methylation signature helps classify VUS [4, 32]. In Malan syndrome, an NFIX variant explains approximately 75% and 19p13.2 deletion approximately 25%, and dosage analysis detects a further 2–5% [13]. OXF recommends EZH2 testing in NSD1-negative Sotos phenotypes [5]. In classical MLC, diagnosis combines clinical and MRI findings (swollen white matter with anterior temporal cysts) with biallelic MLC1 or HEPACAM or a heterozygous GPRC5B variant [8].

5.6. Clinical criteria for MCAP, MPPH and PROS

Source: MCAP PNDS, following Mirzaa 2012 and GeneReviews [22, 41]. Clinical MCAP diagnosis requires megalencephaly or hemimegalencephaly and capillary malformations (two major criteria), and molecular confirmation is necessary for prognosis, management and counselling [22]. Mirzaa 2012 defined MCAP as progressive megalencephaly with vascular abnormalities or syndactyly, and MPPH as progressive megalencephaly with polymicrogyria without vascular abnormalities, syndactyly or heterotopia [41]. Keppler-Noreuil's PROS criteria require PIK3CA mosaicism (in its absence, the diagnosis is presumptive), onset at or shortly after birth and segmental overgrowth, with at least two category A signs (adipose, muscle, nerve or skeletal hypertrophy; vascular malformations; epidermal naevus) or one category B sign (isolated abnormalities, including hemimegalencephaly, megalencephaly or focal cortical dysplasia) [22, 44].

5.7. Report content and interpretation

The report should state gene, reference transcript with version, HGVS description, zygosity, tissue analysed, read depth and allele fraction, and ACMG/AMP classification; ClinGen specifications apply to somatic AKT3, MTOR, PIK3CA and PIK3R2 variants, and laboratories should specify minimum depth, allele fraction threshold, tissue and orthogonal confirmation platform [50]. Blood allele fraction predicts neither severity nor bodily distribution, and a negative blood result does not exclude mosaicism [7, 60]. Parental segregation is useful for the de novo criterion, but leucocyte DNA detects neither all parental mosaicism nor exclusively gonadal mosaicism [3, 50].

5.8. Prenatal diagnosis

In MCAP, prenatal testing is possible in cultured amniotic fluid, because culture selects the mutated clone, or fetal blood, but a negative result does not exclude disease [22]. Prenatal exome sequencing in macrocephalic fetuses yielded 39.1% in a Chinese cohort, with much higher values in non-isolated forms [69]. When the familial variant is known, carrier detection and prenatal or preimplantation diagnosis are possible [1, 3].

6. Differential Diagnosis

6.1. Comparative table

6.2. Benign macrocephaly versus pathological causes

The most frequent practical distinction is between benign familial macrocephaly or BESS and a pathological cause [2]. Large parental HC, normal development, normal neurological and skin examinations and a percentile-parallel HC trajectory support benign disease; HC acceleration, ICH signs, delay, skin macules, asymmetry and vascular malformations require investigation [1, 2]. BESS differs from subdural collection through the cortical vein sign and craniocortical width below 10 mm [2]. OXF's caveat that PTEN and PTCH1 may present with few signs warrants examining proband and parental skin (lipomas, penile macules, trichilemmomas, papillomas) before concluding benignity [1]. A large fontanelle with markedly increased HC leads to the same approach; ERF-related craniosynostosis is associated with macrocephaly, and neonatal progeroid syndromes such as Wiedemann-Rautenstrauch involve relative macrocephaly and intrauterine growth restriction [77].

6.3. Differential diagnosis within the PI3K-AKT-mTOR pathway

MCAP and MPPH share megalencephaly, polymicrogyria and ventriculomegaly, but only MCAP has capillary malformations, syndactyly, segmental overgrowth, cerebral asymmetry and connective tissue dysplasia [3, 22]. However, AKT3 variants may cause laxity and capillary malformations blurring this boundary [3]. Smith-Kingsmore differs through more severe ID, pigmentary mosaicism with somatic variants, hyperphagia and absence of consistent vascular malformations [14, 22]. PHTS differs through mucocutaneous lesions, polyps and cancer predisposition, and rarely has significant focal overgrowth; SOLAMEN syndrome results from a second somatic PTEN alteration on a germline variant [7, 72]. Proteus syndrome begins postnatally and combines cerebriform naevi with characteristic cutaneous, adipose and skeletal overgrowth [1, 7]. Recessive STRADA disease has early lethality and a uniformly poor neurological prognosis [3]. CLOVES may also be associated with hemimegalencephaly [5].

6.4. Epigenetic overgrowth syndromes (summary)

In Sotos syndrome, macrocephaly is the rule and facial gestalt the main clue, most recognisable in early childhood [4, 5]. Weaver is often misdiagnosed as Sotos; round face with hypertelorism, “stuck-on” chin with a horizontal crease, camptodactyly and hoarse cry allow distinction between 1 and 3 years, and macrocephaly is significant in most but not all [1, 4]. Malan has less overgrowth than Sotos, marfanoid traits, more severe ID and more frequent ASD [39]. In TBRS, macrocephaly is present in only half and overweight is the most useful clue [12]. Ventricular dilatation without ICH in Sotos should not be treated, because so-called “arrested hydrocephalus” is not obstructive [4].

6.5. Metabolic causes and leukodystrophies (summary)

Megalencephaly combined with white matter abnormalities, regression, organomegaly or coarse facial features points towards metabolic disease [2]. Canavan is more prevalent in Ashkenazi Jews, with carrier frequency above 1/40 and two founder variants, p.(Glu285Ala) accounting for 80–85% of alleles; NAA is increased in CSF, blood and urine [1, 58]. Alexander disease has radiological diagnostic criteria [58]. In MLC, dissociation between very abnormal MRI and mild physical signs is characteristic, and temporal and frontal cysts are characteristic but not diagnostic [58]. NuRD complex defects (CHD3, CHD4, GATAD2B) and Golgi disorders (HERC, STRADA), as well as TBCK and RAC1, appear in Saudubray's metabolic list of macrocephaly causes [6].

7. Treatments and Risk Management Strategies

7.1. General principles

No available treatment reduces megalencephaly or reverses cortical malformations; management is symptomatic, multidisciplinary and complication-directed, particularly hydrocephalus, tonsillar ectopia, epilepsy, oromotor dysfunction, endocrinopathies and neurodevelopment [3, 7, 22]. PI3K-targeted therapy (alpelisib) is indicated only for severe PROS manifestations and has not demonstrated efficacy for MCAP neurological manifestations; no PI3K or mTOR inhibitor has an approved indication for PIK3R2-, AKT3- or CCND2-related MPPH in any jurisdiction [7, 24, 49]. Benign familial macrocephaly and BESS require no treatment; management is conservative with HC surveillance [2].

7.2. Hydrocephalus and tonsillar ectopia

In MPPH and MCAP, neurosurgical assessment is indicated with rapidly increasing HC, obstructive hydrocephalus, ICH signs or progressive or symptomatic tonsillar ectopia or Chiari malformation [3, 7]. Options include ventriculoperitoneal or ventriculoatrial shunting, endoscopic third ventriculostomy, posterior fossa decompression for symptomatic Chiari type 1 or syringomyelia, and resection of compressive perimedullary masses [7, 22]. Early hydrocephalus treatment may reduce progressive tonsillar ectopia risk, but data are lacking to define the best surgical approach; the observation that MCAP hydrocephalus may be more successfully treated by third ventriculostomy is the GeneReviews author's personal observation, not a comparative result [3, 7]. Families should learn to recognise ICH signs [22]. In Sotos, ventricular dilatation without ICH does not warrant shunting because so-called “arrested hydrocephalus” is not obstructive; only suspected ICH warrants neurosurgical assessment [4]. In Weaver, ventriculomegaly generally had normal CSF pressure without shunting [11].

7.3. Epilepsy

Antiepileptic drugs (AED) are selected according to seizure type by an experienced neurologist; no AED has demonstrated MPPH-specific efficacy, and the French PNDS specifies no preferred class in MCAP [3, 22]. One patient with an AKT3 variant and refractory infantile spasms responded to the ketogenic diet; this isolated report does not allow presenting response as expected [3, 49]. PNDS proposes the ketogenic diet after at least two AED fail and specialist-centre epilepsy surgery for drug resistance (failure of two strategies) [22]. With focal imaging, electroencephalographic (EEG) or semiological findings, hemispherectomy or focus resection should be considered, particularly in hemimegalencephaly and focal cortical dysplasia; severity depends on malformation extent, variant type and mosaicism level [7]. In PHTS, AED follow seizure type and mTOR inhibitors are not first-line epilepsy therapy because of absent evidence [10, 66]. In MLC, epilepsy is generally well controlled with AED, but status epilepticus is relatively frequent [8].

7.4. Feeding, endocrinology and growth

Oromotor dysfunction requires specialist swallowing assessment, dietary modification, speech and language therapy and, when severe, gastrostomy; MCAP feeding difficulties are multifactorial (hypotonia, gastro-oesophageal reflux) and tube feeding is rarely needed [3, 7]. PNDS also mentions proton pump inhibitors, laxatives and transdermal scopolamine for drooling [22]. Hypoglycaemia is treated with intravenous glucose, sugary drinks and meals and cornstarch, with glucagon for persistent hypoglycaemia, and requires assessment of GH and hypothalamic–pituitary–adrenal axes [7]. In GH deficiency, a therapeutic trial may be considered with careful linear growth and overgrowth follow-up; GeneReviews suggests deferring GH until after 2 years, avoiding the main brain growth period, and acknowledges insufficient risk–benefit evidence [7]. Hypothyroidism is more likely in MCAP or PROS with brain involvement and is treated conventionally [7].

7.5. Neurodevelopment, rehabilitation and orthopaedics

Early intervention, physiotherapy (to prevent contractures, scoliosis and hip dislocation), occupational therapy, speech and language therapy and augmentative and alternative communication, which does not impair speech acquisition, are the basis of delay management [3]. In PHTS, the international consensus recommends developmental and full neuropsychological assessment at diagnosis, hearing assessment and speech and language therapy by 3.5 years for language delay, ASD behavioural interventions, depression and anxiety screening from 6 years and annually at least from 12, and first-line sleep hygiene with assessment for obstructive sleep apnoea and ferritin in restless legs syndrome (oral iron for 2–3 months if below 50–75 µg/L) [66]. Orthopaedics treats scoliosis (bracing, arthrodesis), limb length discrepancy (epiphysiodesis, osteotomies) and syndactyly or polydactyly according to function [3, 22]. Vascular malformations are treated with sclerotherapy, laser or oral sirolimus [7]. Social support, care coordination and, when necessary, palliative care and home nursing complete the plan [3]. In MLC, treatment comprises physiotherapy, speech and language therapy and special education, and contact sports and head trauma should be avoided [8].

7.6. Alpelisib (Vijoice®)

Mechanism and indication. Alpelisib is a PI3K inhibitor with predominant α-isoform (p110α) activity; in an inducible CLOVES mouse model it prevented and improved abnormalities, an effect reversed after withdrawal [17]. In the United States it is indicated in adults and children aged 2 or older with severe PROS manifestations requiring systemic therapy, under accelerated approval based on response rate and duration, with continued approval contingent on verification of clinical benefit in confirmatory trials [17]. The European Union indication covers adults and children aged 2 or older with severe or life-threatening PROS manifestations requiring systemic therapy, and treatment should be initiated by a PROS-experienced physician [18]. Efficacy for PROS neurological manifestations such as MCAP is unknown, and the French PNDS considers brain penetration probably incomplete [7, 22]. The indication does not extend to MPPH [24].

Dosing. US and European Union labels have identical doses.

Sources: [7, 17, 18]. Take at the same time with food; a missed dose may be taken within 9 hours, and a vomited dose is not repeated [17, 18]. Tablets (50, 125 and 200 mg in the USA) are swallowed whole; a water suspension may be given orally, by nasogastric tube or gastrostomy; 50 mg granules are only for a 50 mg dose, and multiple sachets must not be combined for 125 or 250 mg or tablets combined with granules [17]. GeneReviews writes “50 gm/day” for dose reduction, a typographical error corresponding to 50 mg daily [7].

Adverse effects and monitoring. In the FDA safety analysis of 57 EPIK-P1 patients (median age 14 years, 2–50; 95% exposed ≥6 months and 79% >1 year), serious reactions occurred in 12% (dehydration and cellulitis), interruption in 11% and dose reduction in 5%; reactions in ≥5% were diarrhoea (16%), stomatitis (16%), hyperglycaemia (12%), eczema and dry skin (7% each), alopecia, headache and cellulitis (5% each); laboratory abnormalities included corrected hypocalcaemia in 60%, hypophosphataemia in 59% (grade 3 in 5%) and increased glucose in 56% (grade 3 in 11%), without grade 4 abnormalities [17]. Warnings include severe hypersensitivity (anaphylaxis, angioedema), severe skin reactions (Stevens-Johnson syndrome, erythema multiforme, toxic epidermal necrolysis, drug reaction with eosinophilia and systemic symptoms, described in oncology), severe hyperglycaemia (hyperosmolar state and fatal ketoacidosis in oncology), pneumonitis, diarrhoea or colitis with dehydration and acute kidney injury, and embryofetal toxicity [17]. Before starting, measure fasting glucose and glycated haemoglobin (HbA1c) and optimise glycaemia; subsequently measure fasting glucose at least weekly for the first 2 weeks and then at least every 4 weeks, and HbA1c every 3 months [17, 18]. The European summary of product characteristics (SmPC) recommends hyperglycaemia specialist consultation with prediabetes, fasting glucose >250 mg/dl (13.9 mmol/l), body mass index ≥30 or age ≥75 years, and, if hyperglycaemia occurs on antidiabetics (including metformin from 10 years or sodium–glucose cotransporter 2 inhibitors), weekly glucose for 8 weeks and then fortnightly [18]. Grade 1–2 rashes are treated with topical corticosteroid and oral antihistamine without dose modification; grade 3 requires interruption; confirmed severe skin reaction, severe hypersensitivity or pneumonitis requires permanent discontinuation; prophylactic antihistamines may reduce rash incidence and severity [17]. Strong CYP3A4 inducers, reducing exposure, and breast cancer resistance protein (BCRP) inhibitors, increasing exposure, should be avoided [17]. Long-term effects on children's growth and development are unknown (rats showed growth plate thickening and dentine thinning), so growth and development should be regularly monitored [17, 18]. The French PNDS considers hypersensitivity, pregnancy, fasting glucose ≥140 mg/dl (≥7.7 mmol/l) and HbA1c >6.4% contraindications to PI3K inhibitors and always prefers clinical trial inclusion [22].

7.7. Comparative regulatory status

The sources consulted do not include information on PROS alpelisib reimbursement or access in Portugal or Brazil. Before European authorisation, the drug was used in France through named-patient temporary authorisation, reserved for severe forms with life-threatening risk or planned mutilating surgery [22]. A patient under 2 years received it off-label through a Novartis managed access programme (section 10) [62].

7.8. Trials and clinical evidence

These studies do not allow indirect comparison: EPIK-P1 is a retrospective review of expanded access patients, mainly CLOVES, counting those without week 24 assessment as non-responders, whereas EPIK-P2 included patients of any severity with a 16-week controlled period considered too short by CHMP [17, 79]. The 27–37.5% response rate refers to ≥20% volumetric reduction in target lesions, not neurological outcomes; the MCAP subgroup had no responses (0/3, very small n), and the evaluable brain lesion (n=1) did not respond either [79]. The only controlled alpelisib trial was negative for its primary objective, and mean volume reductions were below 10% [79]. A systematic review of 17 publications with 114 patients treated outside trials (68.4% paediatric) described clinical improvement in 111 (97.3%) and radiological response in 26 of 60 evaluable patients (the abstract states 47.3%, although 26/60 equals 43.3%), with adverse effects in 56.1%; data derive from small series with heterogeneous outcomes and possible publication bias, and some authors are Novartis trial investigators [85]. The European public assessment report (EPAR) also records, as patient and expert testimony rather than cohort data, that seizures and delay predominate in dysplastic megalencephaly, hemimegalencephaly and focal cortical dysplasia, and brain growth appears greatest in the first two years [79].

7.9. Off-label mTOR inhibitors

Sirolimus (Rapamune®) and everolimus (Afinitor®) have no neurological indication in the genetic megalencephalies; everolimus is approved only for tuberous sclerosis (subependymal giant cell astrocytoma, renal angiomyolipoma and focal seizures) and in oncology [16, 24]. In PROS, low-dose sirolimus modestly reduced the volume of affected tissues (−7.2%), with related adverse effects in 72% of participants and withdrawal for adverse effects in 18%, which requires individual risk–benefit assessment [82]. The alpelisib EPAR reports, without citing a primary source, a modest reduction of pain, pruritus and oedema with sirolimus, poor efficacy in severe forms and about 20% non-responders; in Venot's mouse model, rapamycin improved survival but not the organ abnormalities [79, 80]. In Smith-Kingsmore syndrome, seven patients received off-label sirolimus at doses of 0.5–2 mg daily, with parent-reported improvements in speech, attention, self-injury and hyperphagia; sleep–wake cycle disturbances appeared with trough levels above 3 ng/mL, which led to targeting levels below 3 ng/mL; this is an uncontrolled observation that GeneReviews does not endorse [14, 16]. A 12-year-old boy with megalencephaly and pigmentary mosaicism due to MTOR NM_004958.4:c.7238G>T p.(Ser2413Ile) (23% of cells in affected skin) received sirolimus 1.3 mg twice daily for severe behavioural disturbance, without improvement at 6 months on family and neuropsychological scales; the drug was discontinued, and the three genotyped patients treated previously had conflicting results [86]. In the mouse model of MTOR focal cortical dysplasia, rapamycin suppressed cytomegalic neurons and seizures, which is preclinical evidence [16].

In PHTS, an open-label phase II trial (NCT00971789) described improvement of clinical manifestations with sirolimus in adults during the trial, and Smith reports good rapamycin response in one patient with an extensive arteriovenous malformation [10, 67]. The randomised everolimus trial was negative for the primary neurocognitive outcome, although EEG differences (central alpha and beta power) suggest target engagement [83]. Based on this evidence, the international PHTS consensus recommends against routine mTOR inhibitors for neuropsychiatric or neurocognitive symptoms outside trials (strong recommendation, 100% agreement) and does not consider them first-line for epilepsy; the cancer and overgrowth consensus does not recommend them for colon cancer chemoprevention or polyps except in PTEN-related juvenile polyposis of infancy [66, 87]. AACR notes that PHTS vascular malformations and hamartomas have historically responded to mTOR inhibitors [51].

7.10. Developments in the last 12 months and research

SESAM (NCT05577754, CHU Dijon, with partial funding and drug from Novartis) is the only MCAP-targeted trial: phase II with 6 double-blind months against placebo followed by open-label treatment, alpelisib 250 mg daily in adults or 50 mg daily in children, 20 patients aged 2–40 years with a neurodevelopmental disorder; the primary outcome is a ≥4-point Vineland Adaptive Behavior Scales (VABS-II) gain at 24 months, compared with a theoretical proportion of 6%, and secondary outcomes include quality of life, epilepsy, hypotonia, MRI volumetry and blood–brain barrier passage through optional lumbar puncture; enrolment ended in April 2024 and primary completion is estimated for March 2027 [64]. The trial rationale is that MCAP alpelisib data derive only from compassionate use without demonstrated neurocognitive benefit, and blood–brain barrier crossing is unknown [64]. Other studies: CYH33, a selective PI3Kα inhibitor, in phase I/II in PROS and vascular malformations (China and Japan, 141 planned; NCT06975618); miransertib, an AKT inhibitor, had its PROS and Proteus study terminated (n=50) and retains an active non-recruiting extension (NCT04980872, 5–35 mg/m² daily); taselisib had a terminated study (n=19); EMeRiT (Assistance Publique–Hôpitaux de Paris, NCT07142772) aims to study tumour-risk megalencephaly aetiologies through genome sequencing in 200 children with HC ≥+3 SD and has not yet started recruitment; the PTEN natural history study with ASD (NCT02461446) and a paediatric PHTS registry (NCT06462430) continue [10, 94–96]. In registries, the sirolimus social skills trial in PTEN carriers (NCT06080165) was withdrawn without enrolment [97]. The absence of a US alpelisib approval change through 9 October 2026 follows consultation of Drugs@FDA and the current label and does not exclude non-public regulatory processes [17].

7.11. PHTS: guideline recommendations relevant to the macrocephalic child

The PHTS cancer surveillance recommendations that change the paediatric approach and the divergences between guidelines are summarised here. The 2026 ERN GENTURIS guideline was developed by 17 experts from 9 countries and 3 patient representatives, with a modified Delphi process extended to 48 experts, and issued 17 recommendations with 72–98% agreement [19]. Cumulative risks at 70 years (Hendricks 2023), lower than previously estimated because of ascertainment bias, are 60–80% for female breast cancer, 17–29% for thyroid, 3–9% for kidney, 5–15% for colorectal and 6–12% for melanoma, depending on whether index cases are excluded or included [19].

The divergence in the starting age of thyroid ultrasound reflects the rarity of carcinoma before 10 years (differentiated carcinoma in 3–7% before 20 years and 15% before 30; the earliest at 7 years), the favourable paediatric prognosis and the risk of overdiagnosis [19, 51]. In adulthood, the 2026 ERN GENTURIS guideline recommends annual breast MRI from 25 years and mammography from 40, risk-reducing mastectomy offered from 25 years, annual endometrial ultrasound from 35–40 years, baseline renal imaging and baseline colonoscopy at 35–40 years and baseline dermatological examination at 30 years [19].

7.12. Agents and situations to avoid

In PHTS, caution with diagnostic and therapeutic radiation is recommended because of second neoplasm risk, and skin lesion excision only for suspected malignancy or symptoms because of keloid and recurrence risks; standard cancer treatment, including radiotherapy and chemotherapy, should not be withheld when indicated [10, 66, 87]. In MLC, contact sports and head trauma should be avoided because minor trauma may precipitate status epilepticus or acute motor deterioration [8, 29]. With alpelisib, strong CYP3A4 inducers and BCRP inhibitors are avoided, and treatment is not started in patients with previous severe skin reactions [17, 18]. In BESS, subdural haematoma is the most serious complication, and suspected abusive trauma is an imaging indication in Sampson's algorithm [2].

7.13. Pregnancy, breastfeeding, perioperative care and transition

Alpelisib has embryofetal toxicity: women of reproductive potential should use effective contraception, and men with a pregnant partner or a partner of reproductive potential should use condoms during treatment and for at least one week afterwards; the PNDS considers pregnancy a contraindication to PI3K inhibitors [17, 18, 22]. It is not known whether alpelisib passes into breast milk; the European SmPC recommends not breastfeeding during treatment and for at least 1 week after the last dose [18]. In a pregnancy with a fetus with Weaver syndrome, a large newborn should be anticipated, and in Sotos syndrome maternal pre-eclampsia occurs in about 15% [4, 11]. In the perioperative period of patients with PROS and extensive venous malformations, the thrombotic risk warrants assessment of coagulation and D-dimers, and haematological surveillance for thrombosis or coagulopathy is recommended after surgery [7, 22]. Transition to adult services takes place in late adolescence, maintaining neurological, orthopaedic and, in PHTS, cancer surveillance [19, 22].

7.14. Epigenetic overgrowth syndromes: treatment specifics

In Sotos, treatment addresses complications (learning difficulty, behaviour, heart defects, kidney abnormalities, scoliosis, seizures), and ventricular dilatation without ICH is not treated [4]. In EZH2 overgrowth, a low threshold to investigate tumour-suggestive symptoms is recommended, particularly neuroblastoma and haematological neoplasms, and in TBRS a low threshold for full blood count; systematic screening divergences appear in section 11 [11, 12]. In TBRS, aortic root surveillance is determined by diameter and cardiologist, and in Malan aortic root dilatation has been described in four patients, one with dissection at 38 years [12, 13].

8. Reproductive Options and Family Counselling

8.1. Entity-specific recurrence risks

8.2. Parental assessment and cascade testing

In MPPH, parents should undergo variant testing and baseline neurological assessment with HC measurement, because a parent with somatic and germline mosaicism may be mildly or minimally affected; leucocyte DNA detects neither all mosaicism nor exclusively gonadal mosaicism [3]. In PHTS, parents of a simplex case should be tested even when clinically unaffected because expression may be subtle; an untested clinically unaffected parent over 30 years has a low but increased carrier risk [10]. In all AD forms, parental HC measurement is part of examination [1].

8.3. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing are possible when the familial variant is known [1, 3]. In MCAP, because the variant is generally postzygotic and not inherited, prenatal diagnosis is not usually indicated given procedure-related fetal loss risk but may be performed at parental request with specialist ultrasound; in rare constitutional cases, chorionic villus sampling or amniocentesis may be offered; in an affected parent's pregnancy, prenatal diagnosis or ultrasound surveillance is possible [22]. A negative prenatal result does not exclude MCAP [22]. In Sotos, prenatal variant identification does not predict severity [4]. Without a molecular diagnosis, proband DNA banking is recommended [3]. In France, prenatally diagnosed MCAP is eligible for medical termination under French law [22].

8.4. Counselling minors

In PHTS, familial variant testing in early childhood is often appropriate because early detection guides developmental assessment and management [10, 51]. In other AD syndromes, diagnosing an at-risk child has immediate clinical utility (hydrocephalus, neurodevelopmental, Chiari surveillance) and does not raise the issue of late-onset predictive testing [3, 7].

9. Special Considerations: Population Diversity and Epidemiology

9.1. What each number measures

Orphanet MCAP and MPPH prevalence classes derive from the same published series as GeneReviews and are not independent estimates [24]. Sources provide no TBRS, Weaver or Smith-Kingsmore prevalence because mild forms evade diagnosis [11, 14]. The sources consulted provide no Portuguese or Brazilian epidemiological data for these entities, which does not demonstrate absence of publications in those countries.

9.2. Founder populations and geographical differences

In Japan, the 1.9 Mb 5q35 microdeletion is the most frequent finding in Sotos syndrome, whereas intragenic variants predominate in Europe and North America; this difference determines the choice of first test (microarray or sequencing with dosage analysis) [4, 32–34]. Canavan disease is more prevalent in Ashkenazi Jews, with a carrier frequency above 1/40 and two founder variants [1]. In MLC, MLC1 (NM_015166.4) founder variants are known in the Agrawal community of India (c.135dup), in Libyan and Turkish Jews (c.176G>A, with a higher-than-expected carrier frequency), in Japanese people (c.278C>T), in Koreans (c.824C>A, 70% of alleles) and in Egyptians (c.908_918delinsGCA); no founder population has been described for HEPACAM [8, 29, 57]. The Korean MCAP series showed a spectrum of PIK3CA variants without hotspot predominance, with p.(Gly914Arg) in 4 of 12 patients [60]. The PHTS cohort of the international consensus guidelines is mainly of European ancestry, with an expert panel almost entirely from the USA, which limits generalisability [66].

9.3. Special populations

Extremely premature infants with macrocephaly pose a particular problem: intraventricular haemorrhage and posthaemorrhagic hydrocephalus may mask primary megalencephaly, illustrated by the case in section 10 [62]. In adolescents and adults, Sotos and TBRS gestalt changes with age and height may normalise while macrocephaly persists; late PHTS diagnosis, frequently only in adulthood, reinforces the value of childhood HC and z-score measurement [4, 12, 70]. In PHTS, motor, executive and memory deficits may emerge in adulthood even without childhood delay [10].

9.4. European reference networks and patient organisations

The relevant ERN are ERN ITHACA (rare malformations and ID, including Sotos, Weaver, TBRS and MCAP/MPPH; https://ern-ithaca.eu), VASCERN (vascular anomalies, including PROS; https://vascern.eu) and ERN GENTURIS (PHTS; https://www.genturis.eu), author of the 2026 guideline, whose Delphi process included PTEN Italia [19]. In France, MCAP is followed by reference centres in the AnDDI-Rares, DéfiScience and FAVA-Multi networks [22]. Patient organisations include M-CM Network (https://m-cm.net) and M-CM France for MCAP, CLOVES Syndrome Community (https://clovessyndrome.org) for PROS, PTEN Foundation (https://ptenfoundation.org), the Sotos syndrome support association (https://www.sotossyndrome.org), the UK's Child Growth Foundation (https://childgrowthfoundation.org) and TBRS Community (https://tbrsyndrome.org) [1, 5, 22]. Addresses were accessed on 9 October 2026.

10. Illustrative Clinical Case

Anderson and Velinov published a case in 2025 of a female infant born at 26 weeks and 6 days' gestation, with grade IV intraventricular haemorrhage, severe communicating hydrocephalus, corpus callosum agenesis and limited sulcation with a thick cortical mantle [62]. She required ventriculoperitoneal shunting with several malfunctions, external ventricular drainage and revisions, and had seizures; she had dysmorphism (long face, parietal bossing, facial asymmetry, mild hypotelorism, low-set ears), a two-vessel umbilical cord, nystagmus, cardiomegaly, central venous thrombosis and a back haemangioma, and depended on gastrostomy and oxygen [62]. An overgrowth and macrocephaly panel identified mosaic PIK3CA NM_006218.4:c.1093G>A p.(Glu365Lys), as well as a TGFB3 VUS (c.730G>A p.(Glu244Lys)) and an approximately 379 kb 4p16.1 duplication also of uncertain significance; Beckwith-Wiedemann and Silver-Russell methylation testing was normal [59, 62]. She was assessed in genetics at 8 months (5 months corrected) and started alpelisib at 10 months (7 corrected), 25 mg daily (half a 50 mg tablet), through a Novartis managed access programme, off-label because she was under 2 years; she was the third and youngest patient under 2 years in the programme [62].

At 11 months (8 corrected), she had severe generalised hypotonia, megalencephaly with frontal bossing (HC 49.5 cm, >99th percentile) and an involuting left scapular haemangioma, documented in article Figure 1 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12503962/figure/fig1/) [62]. After approximately one year of treatment, at 19 months (16 corrected), neurological and neurosurgical stabilisation occurred, with fewer seizures and shunt malfunctions; frontoparietal polymicrogyria, Chiari malformation and corpus callosum agenesis persisted, and HC was 52 cm (>99th percentile) [62]. Suspected pneumonitis was excluded by radiography, one hypoglycaemic episode was attributed to hyperinsulinism or late dumping syndrome after gastrostomy, IGF-1 was low and no adverse effects were attributed to the drug; development remained markedly delayed without verbal language, and the child entered palliative care in a long-term care unit [62]. The authors concluded that there was no robust cognitive improvement and the drug does not reverse prior lesions; prematurity and intraventricular haemorrhage confound assessment, and no standardised measures were used [62].

The case illustrates three points: genetic megalencephaly may be obscured by an acquired hydrocephalus cause; PIK3CA mosaicism was diagnosed by panel, which does not always occur in blood; and alpelisib use below 2 years at half-dose is outside US and EU labels and trial evidence, which has not demonstrated MCAP neurological benefit [17, 18, 62, 79]. The reported management is not recommended practice.

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

In benign familial macrocephaly and BESS, prognosis is excellent, with resolution of the enlargement by 3 years [2]. In MCAP, most patients have mild to moderate delay with continuous progress and fewer than 10% severe disability; severity is mainly neurological and does not clearly correlate with the brain abnormalities [7, 22]. Excessive head growth is concentrated in the first two years, the period of highest risk of obstructive hydrocephalus and tonsillar ectopia; the PNDS recommends regular neuroradiological monitoring in the first 8 years, particularly the first 2 [3, 22]. In MPPH, ID is almost universal and more severe with extensive polymicrogyria and neonatal-onset or generalised epilepsy; CCND2 variants are associated with the most severe phenotypes [3]. In Smith-Kingsmore syndrome, survival into adulthood is common [14]. In PHTS, long-term prognosis depends on cancer surveillance [19]. In Sotos syndrome, height may normalise in adulthood, but macrocephaly persists, and adults may have depression, anxiety and isolation [4, 32]. In classical MLC, loss of independent walking generally occurs in the first or second decade, with great variability; some patients die in adolescence and others live into adulthood [29, 58]. Infantile Alexander disease leads to death in the first decade [1].

11.2. Tumour surveillance: source divergences

11.3. Surveillance table

12. Summary and Final Considerations

Macrocephaly is defined by HC >2 SD, but pathological causes predominate only above +3 SD, and the first task in a macrocephalic child is to exclude ICH and hydrocephalus. Parental HC measurement, the HC curve expressed in SD, its relationship to height and development, and careful skin examination distinguish benign familial macrocephaly and BESS, requiring no genetic testing, from forms warranting investigation.

Neuroimaging, preferably neuroradiologist-reviewed MRI, separates secondary causes, metabolic and developmental megalencephaly, and guides testing. In a child with delay, ID or congenital abnormalities, exome or genome sequencing with copy number variant analysis is first-line; PTEN should be tested with dosage and promoter analysis when macrocephaly accompanies ASD, delay, skin lesions, vascular malformations or polyps; and MCAP and focal megalencephalies require deep sequencing of skin, saliva or brain tissue because blood is frequently negative and a negative result does not exclude diagnosis.

PI3K-AKT-mTOR pathway megalencephalies share a pathway activation mechanism, with cyclin D2 as a convergence point, and a critical first-two-year period during which serial MRI allows hydrocephalus and tonsillar ectopia treatment before decompensation. MCAP differs from MPPH through capillary malformations, syndactyly, asymmetry and mosaicism, bringing recurrence risk close to population risk, whereas MPPH is AD with possible transmission. In Sotos, Weaver and TBRS, macrocephaly is part of global overgrowth and diagnosis rests on gestalt, bone age and molecular testing; Sotos ventricular dilatation does not require shunting.

Alpelisib has been indicated in the USA since 2022 and, conditionally, in the European Union since July 2026, for severe manifestations of PROS, on the basis of a retrospective study; the only controlled trial missed its primary endpoint, the MCAP subgroup had no radiological responses and efficacy on the brain remains unknown, so the SESAM trial, with completion expected in 2027, will be decisive. mTOR inhibitors have no place in routine neurocognitive management of PHTS. Cancer surveillance divides the sources: screening is agreed in PHTS and in Beckwith-Wiedemann, but Wilms tumour screening in PROS and neuroblastoma screening in EZH2 overgrowth receive opposite recommendations depending on the risk threshold adopted.

13. References

1. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 224–6 (Macrocephaly).
2. Accogli A, Geraldo AF, Piccolo G, Riva A, Scala M, Balagura G, et al. Diagnostic approach to macrocephaly in children. Front Pediatr. 2022;9:794069. doi:10.3389/fped.2021.794069
3. Mirzaa G. MPPH Syndrome. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2016 Nov 17; last update 2022 Jul 28. Available from: https://www.ncbi.nlm.nih.gov/books/NBK396098/
4. Ocansey S, Cole TRP, Rahman N, Tatton-Brown K. Sotos Syndrome. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2004 Dec 17; last update 2025 Jun 5. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1479/
5. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 272–5 (Overgrowth).
6. Saudubray JM, García-Cazorla Á. Clinical Approach to Inborn Errors of Metabolism in Paediatrics. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, sections 1.5.5 (Table 1.22) and 1.5.6. doi:10.1007/978-3-662-63123-2_1
7. Mirzaa G, Graham JM Jr, Keppler-Noreuil K. PIK3CA-Related Overgrowth Spectrum. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2013 Aug 15; last update 2021 Dec 23 (review 2023 Apr 6). Available from: https://www.ncbi.nlm.nih.gov/books/NBK153722/
8. Min R, Abbink TEM, van der Knaap MS. Megalencephalic Leukoencephalopathy with Subcortical Cysts. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2003 Aug 11; last update 2023 Jul 27. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1535/
9. GeneGuidance. Conditions retrieved by search term (macrocephaly/megalencephaly/overgrowth) — condition records with identifiers OMIM/Orphanet.
10. Tramontana T, Yehia L. PTEN Hamartoma Tumor Syndrome. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2001 Nov 29; last update 2026 Jul 28. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1488/
11. Ocansey S, Tatton-Brown K. EZH2-Related Overgrowth. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2013 Jul 18; last update 2024 Mar 21. Available from: https://www.ncbi.nlm.nih.gov/books/NBK148820/
12. Ostrowski PJ, Tatton-Brown K. Tatton-Brown-Rahman Syndrome. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2022 Jun 30. Available from: https://www.ncbi.nlm.nih.gov/books/NBK581652/
13. Priolo M. NFIX-Related Malan Syndrome. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2024 Aug 1. Available from: https://www.ncbi.nlm.nih.gov/books/NBK605494/
14. Prada CE, Raski CR, Mirzaa G, Besterman A, Krueger DA, Liu AC. Smith-Kingsmore Syndrome. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2025 Nov 20. Available from: https://www.ncbi.nlm.nih.gov/books/NBK619247/
15. Mitchel MW, Myers SM, Heidlebaugh AR, Taylor CM, Rea H, Neuhaus E, et al. CHD8-Related Neurodevelopmental Disorder with Overgrowth. In: Adam MP, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Initial posting 2022 Oct 27. Available from: https://www.ncbi.nlm.nih.gov/books/NBK585456/
16. GeneGuidance. MTOR — gene record (clinical curation).
17. Novartis Pharmaceuticals Corporation. VIJOICE (alpelisib) tablets and oral granules. Prescribing information. NDA 215039 and NDA 218466. Silver Spring (MD): U.S. Food and Drug Administration; revised 2025 Jul (SPL effective 2025-11-04). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=638069a4-aee0-46c9-9e21-54ad449c163c
18. European Medicines Agency. Vijoice (alpelisib): EPAR — product information (EMEA/H/C/006539). Amsterdam: EMA; 2026. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/vijoice
19. Hoogerbrugge N, Blatnik A, Lautrup CK, Hüneburg R, Turchetti D, van Altena A, et al. ERN GENTURIS cancer surveillance guideline for individuals with PTEN hamartoma tumour syndrome (PHTS). Eur J Hum Genet. 2026 Jul 16. doi:10.1038/s41431-026-02181-z
20. Orphanet. Megalencephaly-capillary malformation-polymicrogyria syndrome (ORPHA:60040). Available from: https://www.orpha.net/en/disease/detail/60040
21. GeneGuidance. PIK3CA — gene record (clinical curation).
22. Olivier-Faivre L, Thauvin C, Guibaud L, Bahi-Buisson N, coordinators. Protocole National de Diagnostic et de Soins (PNDS): Syndrome MCAP (Megalencephaly-Capillary malformation-Polymicrogyria). Saint-Denis: Haute Autorité de Santé; 2021 Sep. Available from: https://www.has-sante.fr/upload/docs/application/pdf/2021-10/pnds-mcap_sept_2021.pdf
23. Orphanet. Megalencephaly-polymicrogyria-postaxial polydactyly-hydrocephalus syndrome (ORPHA:83473). Available from: https://www.orpha.net/en/disease/detail/83473
24. GeneGuidance. PIK3R2 — gene record (clinical curation).
25. Orphanet. Hemimegalencephaly (ORPHA:99802). Available from: https://www.orpha.net/en/disease/detail/99802
26. Orphanet. PTEN hamartoma tumor syndrome (ORPHA:306498). Available from: https://www.orpha.net/en/disease/detail/306498
27. Orphanet. Bannayan-Riley-Ruvalcaba syndrome (ORPHA:109). Available from: https://www.orpha.net/en/disease/detail/109
28. Orphanet. Megalencephalic leukoencephalopathy with subcortical cysts (ORPHA:2478). Available from: https://www.orpha.net/en/disease/detail/2478
29. GeneGuidance. MLC1 — gene record (clinical curation).
30. Orphanet. Isolated megalencephaly (ORPHA:2477). Available from: https://www.orpha.net/en/disease/detail/2477
31. GeneGuidance. NSD1 — gene record (condition mapping Sotos 1 ↔︎ ORPHA 1627, incorrect).
32. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 202–4 (Sotos syndrome).
33. Kurotaki N, Imaizumi K, Harada N, Masuno M, Kondoh T, Nagai T, et al. Haploinsufficiency of NSD1 causes Sotos syndrome. Nat Genet. 2002;30(4):365-6. doi:10.1038/ng863
34. Douglas J, Hanks S, Temple IK, Davies S, Murray A, Upadhyaya M, et al. NSD1 mutations are the major cause of Sotos syndrome and occur in some cases of Weaver syndrome but are rare in other overgrowth phenotypes. Am J Hum Genet. 2003;72(1):132-43. doi:10.1086/345647
35. Gibson WT, Hood RL, Zhan SH, Bulman DE, Fejes AP, Moore R, et al. Mutations in EZH2 cause Weaver syndrome. Am J Hum Genet. 2012;90(1):110-8. doi:10.1016/j.ajhg.2011.11.018
36. Tatton-Brown K, Hanks S, Ruark E, Zachariou A, Duarte SV, Ramsay E, et al. Germline mutations in the oncogene EZH2 cause Weaver syndrome and increased human height. Oncotarget. 2011;2(12):1127-33. doi:10.18632/oncotarget.385
37. Imagawa E, Higashimoto K, Sakai Y, Numakura C, Okamoto N, Matsunaga S, et al. Mutations in genes encoding polycomb repressive complex 2 subunits cause Weaver syndrome. Hum Mutat. 2017;38(6):637-48. doi:10.1002/humu.23200
38. Tatton-Brown K, Seal S, Ruark E, Harmer J, Ramsay E, Del Vecchio Duarte S, et al. Mutations in the DNA methyltransferase gene DNMT3A cause an overgrowth syndrome with intellectual disability. Nat Genet. 2014;46(4):385-8. doi:10.1038/ng.2917
39. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 206–7 (Malan syndrome).
40. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 732–4 (Megalencephaly-capillary malformation-polydactyly syndrome).
41. Mirzaa GM, Conway RL, Gripp KW, Lerman-Sagie T, Siegel DH, deVries LS, et al. Megalencephaly-capillary malformation (MCAP) and megalencephaly-polydactyly-polymicrogyria-hydrocephalus (MPPH) syndromes: two closely related disorders of brain overgrowth and abnormal brain and body morphogenesis. Am J Med Genet A. 2012;158A(2):269-91. doi:10.1002/ajmg.a.34402
42. Rivière JB, Mirzaa GM, O'Roak BJ, Beddaoui M, Alcantara D, Conway RL, et al. De novo germline and postzygotic mutations in AKT3, PIK3R2 and PIK3CA cause a spectrum of related megalencephaly syndromes. Nat Genet. 2012;44(8):934-40. doi:10.1038/ng.2331
43. Mirzaa G, Parry DA, Fry AE, Giamanco KA, Schwartzentruber J, Vanstone M, et al. De novo CCND2 mutations leading to stabilization of cyclin D2 cause megalencephaly-polymicrogyria-polydactyly-hydrocephalus syndrome. Nat Genet. 2014;46(5):510-5. doi:10.1038/ng.2948
44. Keppler-Noreuil KM, Rios JJ, Parker VE, Semple RK, Lindhurst MJ, Sapp JC, et al. PIK3CA-related overgrowth spectrum (PROS): diagnostic and testing eligibility criteria, differential diagnosis, and evaluation. Am J Med Genet A. 2015;167A(2):287-95. doi:10.1002/ajmg.a.36836
45. Butler MG, Dasouki MJ, Zhou XP, Talebizadeh Z, Brown M, Takahashi TN, et al. Subset of individuals with autism spectrum disorders and extreme macrocephaly associated with germline PTEN tumour suppressor gene mutations. J Med Genet. 2005;42(4):318-21. doi:10.1136/jmg.2004.024646
46. GeneGuidance. CCND2 — gene record (clinical curation).
47. Lamari F, Rossignol F, Mitchell GA. Disorders of Intracellular Triglyceride and Phospholipid Metabolism. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 35, section 35.5 (Table 35.1). doi:10.1007/978-3-662-63123-2_35
48. Kuentz P, St-Onge J, Duffourd Y, Courcet JB, Carmignac V, Jouan T, et al. Molecular diagnosis of PIK3CA-related overgrowth spectrum (PROS) in 162 patients and recommendations for genetic testing. Genet Med. 2017;19(9):989-97. doi:10.1038/gim.2016.220
49. GeneGuidance. AKT3 — gene record (clinical curation).
50. Lai A, Soucy A, El Achkar CM, Barkovich AJ, Cao Y, DiStefano M, et al. The ClinGen Brain Malformation Variant Curation Expert Panel: rules for somatic variants in AKT3, MTOR, PIK3CA, and PIK3R2. Genet Med. 2022;24(11):2240-8. doi:10.1016/j.gim.2022.07.020
51. Schultz KAP, MacFarland SP, Perrino MR, Mitchell SG, Kamihara J, Nelson AT, et al. Update on pediatric surveillance recommendations for PTEN hamartoma tumor syndrome, DICER1-related tumor predisposition, and tuberous sclerosis complex. Clin Cancer Res. 2025;31(2):234-44. doi:10.1158/1078-0432.CCR-24-1947
52. Orphanet. Megalencephaly-severe kyphoscoliosis-overgrowth syndrome (ORPHA:457359). Available from: https://www.orpha.net/en/disease/detail/457359
53. Orphanet. Polyhydramnios-megalencephaly-symptomatic epilepsy syndrome (ORPHA:500533). Available from: https://www.orpha.net/en/disease/detail/500533
54. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 210–1 (Weaver syndrome).
55. GeneGuidance. CHD8 — gene record (clinical curation).
56. Tatton-Brown K, Loveday C, Yost S, Clarke M, Ramsay E, Zachariou A, et al. Mutations in epigenetic regulation genes are a major cause of overgrowth with intellectual disability. Am J Hum Genet. 2017;100(5):725-36. doi:10.1016/j.ajhg.2017.03.010
57. GeneGuidance. HEPACAM — gene record (clinical curation).
58. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 208–11 (Leukodystrophy and leukoencephalopathy).
59. 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
60. Park HJ, Shin CH, Yoo WJ, Cho TJ, Kim MJ, Seong MW, et al. Detailed analysis of phenotypes and genotypes in megalencephaly-capillary malformation-polymicrogyria syndrome caused by somatic mosaicism of PIK3CA mutations. Orphanet J Rare Dis. 2020;15(1):205. doi:10.1186/s13023-020-01480-y
61. Chen WL, Pao E, Owens J, Glass I, Pritchard C, Shirts BH, et al. The utility of cerebrospinal fluid-derived cell-free DNA in molecular diagnostics for the PIK3CA-related megalencephaly-capillary malformation (MCAP) syndrome: a case report. Cold Spring Harb Mol Case Stud. 2022;8(3):a006188. doi:10.1101/mcs.a006188
62. Anderson S, Velinov M. An extremely preterm infant with PIK3CA-related overgrowth spectrum (PROS): alpelisib treatment and outcome. Case Rep Genet. 2025;2025:6839348. doi:10.1155/crig/6839348
63. Garde A, Guibaud L, Goldenberg A, Petit F, Dard R, Roume J, et al. Clinical and neuroimaging findings in 33 patients with MCAP syndrome: a survey to evaluate relevant endpoints for future clinical trials. Clin Genet. 2021;99(5):650-61. doi:10.1111/cge.13918
64. Luu M, Vabres P, Espitalier A, Maurer A, Garde A, Racine C, et al. A phase II double-blind multicentre, placebo-controlled trial to assess the efficacy and safety of alpelisib (BYL719) in paediatric and adult patients with megalencephaly-capillary malformation polymicrogyria syndrome (MCAP): the SESAM study protocol. BMJ Open. 2024;14(12):e084614. doi:10.1136/bmjopen-2024-084614
65. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 216–21 (Lissencephaly, polymicrogyria and neuronal migration disorders).
66. Dhawan A, Liu D, Baitamouni S, Anthony K, Srivastava S, Hardan AY, et al. Neurodevelopmental and neurologic manifestations of PTEN hamartoma tumor syndrome: management recommendations. Neurol Genet. 2025;11(5):e200299. doi:10.1212/NXG.0000000000200299
67. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 742–3 (Bannayan-Riley-Ruvalcaba syndrome).
68. Tovy A, Rosas C, Gaikwad AS, Medrano G, Zhang L, Reyes JM, et al. Perturbed hematopoiesis in individuals with germline DNMT3A overgrowth Tatton-Brown-Rahman syndrome. Haematologica. 2022;107(4):887-98. doi:10.3324/haematol.2021.278990
69. Zhou H, Fu F, Huang R, Yu Q, Yan S, Lu J, et al. Prenatal exome sequencing for fetal macrocephaly: a large prospective observational cohort study. Prenat Diagn. 2026;46(5-6):819-31. doi:10.1002/pd.6818
70. Gupta NA, Shur N, Berger SI. A stepwise approach to macrocephaly: clinical clues to the rare diagnosis of PTEN hamartoma tumor syndrome. Pediatr Ann. 2026;55(4):e139-42. doi:10.3928/19382359-20260309-02
71. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 190–1 (Intellectual disability).
72. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 736–8 (Proteus syndrome).
73. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Glutaric aciduria (type I). In: Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. Chapter 9, p. 70–81.
74. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. D-2-hydroxyglutaric (DL-2-hydroxyglutaric) aciduria. In: Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. Chapter 11, p. 87–94.
75. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 194–7 (Intellectual disability with apparent X-linked inheritance).
76. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 578–81 (Juvenile polyposis syndrome).
77. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 204–5 (Large fontanelle).
78. European Medicines Agency. Withdrawal of application for the marketing authorisation of Vijoice (alpelisib). EMA/492147/2023. Amsterdam: EMA; 2023 Nov 10. Available from: https://www.ema.europa.eu/en/documents/medicine-qa/questions-and-answers-withdrawal-application-marketing-authorisation-vijoice-alpelisib_en.pdf
79. European Medicines Agency. Vijoice: EPAR — public assessment report (EMADOC-1829012207-22240). Amsterdam: EMA; 2026. Available from: https://www.ema.europa.eu/en/documents/assessment-report/vijoice-epar-public-assessment-report_en.pdf
80. Venot Q, Blanc T, Rabia SH, Berteloot L, Ladraa S, Duong JP, et al. Targeted therapy in patients with PIK3CA-related overgrowth syndrome. Nature. 2018;558(7711):540-6. doi:10.1038/s41586-018-0217-9
81. ClinicalTrials.gov. EPIK-P2: study assessing the efficacy, safety and PK of alpelisib (BYL719) in pediatric and adult patients with PROS. NCT04589650. Available from: https://clinicaltrials.gov/study/NCT04589650
82. Parker VER, Keppler-Noreuil KM, Faivre L, Luu M, Oden NL, De Silva L, et al. Safety and efficacy of low-dose sirolimus in the PIK3CA-related overgrowth spectrum. Genet Med. 2019;21(5):1189-98. doi:10.1038/s41436-018-0297-9
83. Srivastava S, Jo B, Zhang B, Frazier T, Gallagher AS, Peck F, et al. A randomized controlled trial of everolimus for neurocognitive symptoms in PTEN hamartoma tumor syndrome. Hum Mol Genet. 2022;31(20):3393-404. doi:10.1093/hmg/ddac111
84. ClinicalTrials.gov. RAD001 and neurocognition in PTEN hamartoma tumor syndrome. NCT02991807. Available from: https://clinicaltrials.gov/study/NCT02991807
85. Pellegrino F, Reynolds G, Cardaropoli S, Luca M, Massuras S, Carli D, et al. Alpelisib in PIK3CA-related overgrowth spectrum (PROS): a systematic review of real-world evidence in over 100 patients. Cells. 2026;15(9):788. doi:10.3390/cells15090788
86. Bonniaud B, Luu M, Cormier C, Racine C, Espitalier A, Malbranche C, et al. Lack of behavioural improvement with sirolimus in a patient with MTOR-related macrocephaly with pigmentary mosaicism: a new case report. Eur J Med Genet. 2025;75:105012. doi:10.1016/j.ejmg.2025.105012
87. Dhawan A, Baitamouni S, Liu D, Yehia L, Anthony K, McCarther A, et al. Cancer and overgrowth manifestations of PTEN hamartoma tumor syndrome: management recommendations from the International PHTS Consensus Guidelines Working Group. Clin Cancer Res. 2025;31(9):1754-65. doi:10.1158/1078-0432.CCR-24-3819
88. ClinicalTrials.gov. EPIK-P4: phase II single-arm study of alpelisib in PROS. NCT06997588. Available from: https://clinicaltrials.gov/study/NCT06997588
89. Relay Therapeutics. Relay Therapeutics announces initial clinical data demonstrating that zovegalisib has potential for differentiated safety and efficacy in patients with PIK3CA-driven vascular anomalies [press release]. Cambridge (MA): Relay Therapeutics; 2026 May 19. Available from: https://www.biospace.com/press-releases/relay-therapeutics-announces-initial-clinical-data-demonstrating-that-zovegalisib-has-potential-for-differentiated-safety-and-efficacy-in-patients-with-pik3ca-driven-vascular-anomalies
90. ClinicalTrials.gov. ReInspire: RLY-2608 (zovegalisib) in PROS and PIK3CA-driven malformations. NCT06789913. Available from: https://clinicaltrials.gov/study/NCT06789913
91. ClinicalTrials.gov. SVG103 (paxalisib) in FCD II, TSC or hemimegalencephaly. NCT07287202. Available from: https://clinicaltrials.gov/study/NCT07287202
92. ClinicalTrials.gov. Everolimus for improving social abilities in PTEN germline mutations. NCT07218575. Available from: https://clinicaltrials.gov/study/NCT07218575
93. ClinicalTrials.gov. SESAM: alpelisib in MCAP. NCT05577754. Available from: https://clinicaltrials.gov/study/NCT05577754
94. ClinicalTrials.gov. Study to evaluate the safety, tolerability, pharmacokinetics and preliminary efficacy of CYH33 in patients with PIK3CA-related overgrowth spectrum (PROS) and PIK3CA-related vascular malformations (PRVM). NCT06975618. Sponsor: Haihe Biopharma. Available from: https://clinicaltrials.gov/study/NCT06975618
95. ClinicalTrials.gov. A study of the safety and tolerability in participants with PIK3CA-related overgrowth spectrum or Proteus syndrome who are being treated with miransertib (MK-7075) in other studies (MK-7075-006). NCT04980872. Sponsor: Merck Sharp & Dohme. Available from: https://clinicaltrials.gov/study/NCT04980872
96. ClinicalTrials.gov. EMeRiT: etiologies of megalencephaly associated with a detectable tumor risk. NCT07142772. Available from: https://clinicaltrials.gov/study/NCT07142772
97. ClinicalTrials.gov. Sirolimus for improving social abilities in people with PTEN germline mutations. NCT06080165. Available from: https://clinicaltrials.gov/study/NCT06080165
98. Tischkowitz M, Colas C, Pouwels S, Hoogerbrugge N; PHTS Guideline Development Group; European Reference Network GENTURIS. Cancer surveillance guideline for individuals with PTEN hamartoma tumour syndrome. Eur J Hum Genet. 2020;28(10):1387-93. doi:10.1038/s41431-020-0651-7
99. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 564–7 (Cowden syndrome (PTEN hamartoma tumour syndrome)).
100. 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
101. Villani A, Greer MC, Kalish JM, Nakagawara A, Nathanson KL, Pajtler KW, et al. Recommendations for cancer surveillance in individuals with RASopathies and other rare genetic conditions with increased cancer risk. Clin Cancer Res. 2017;23(12):e83-90. doi:10.1158/1078-0432.CCR-17-0631