Pallister-Hall and Greig Syndromes (GLI3-Related Disorders)

Spectrum of developmental malformations caused by heterozygous variants in the GLI3 gene, an effector of the hedgehog signalling pathway, which includes Greig cephalopolysyndactyly syndrome, Pallister-Hall syndrome with hypothalamic hamartoma, apparently isolated postaxial and preaxial polydactylies and the contiguous gene deletion forms at 7p14, with a phenotype determined largely by the type and position of the variant.

Literature and regulatory information updated: 17 October 2026.

0. Initial Reference Table

Depending on the class and location of the variant, the GLI3 gene (7p14.1) causes two main syndromic phenotypes and a set of apparently isolated digital phenotypes. Greig cephalopolysyndactyly syndrome (GCPS) results from haploinsufficiency; Pallister-Hall syndrome (PHS) results from truncating variants in the middle third of the gene that produce a constitutive repressor protein [1, 2]. Table 0.1 lists the entities within scope with their identifiers; table 0.2 summarises the data for the first consultation. The broader differential of polydactylies and brachydactylies due to other genes is covered as differential diagnosis in section 6.

Table 0.1 — GLI3-related entities and identifiers

Three identification notes complete the table. OMIM entry 241800, which some databases still designate "hypothalamic hamartomas", is now titled "Pallister-Hall-like syndrome" (PHLS), with autosomal recessive inheritance due to biallelic SMO variants (7q32.1); it is a differential diagnosis and not a GLI3 entity, and non-syndromic hamartoma due to a somatic GLI3 variant has no OMIM number of its own [1, 10, 20]. Orphanet records autosomal recessive inheritance for PAP-A because category ORPHA:93334 aggregates nine OMIM entries for different genes, which does not alter the dominant inheritance of the GLI3 form [11]. The association between GLI3 and isolated tibial hemimelia is an Orphanet attribution, validated by a publication not included in the sources consulted, and does not appear in GeneReviews or in the reference books as a GLI3 phenotype; it is presented here as a recorded association and not as an established allelic disorder [2, 19].

Table 0.2 — Data for the first consultation

1. Introduction

1.1. Greig cephalopolysyndactyly syndrome

Greig cephalopolysyndactyly syndrome is named after David Middleton Greig, a Scottish surgeon who described it in a 1926 manuscript published in the Edinburgh Medical Journal under the title "Oxycephaly"; the eponym should not be confused with that of the composer Grieg [7, 38]. There followed the synopsis of hand malformations by Temtamy and McKusick, in 1969, and the description by Marshall and Smith, in 1970, who used the designation "frontodigital syndrome" for a dominantly inherited disorder with normal intelligence [38]. Smith's Recognizable Patterns of Human Malformation classifies GCPS in the chapter on syndromes with craniosynostosis, with the subtitle "preaxial and postaxial polydactyly, syndactyly, frontal bossing" [38]. The syndrome was defined by the association of polysyndactyly, macrocephaly and hypertelorism, with autosomal dominant transmission, high penetrance and variability between and within families [7, 38]. Biesecker's 2008 review discourages the expression "Greig syndrome" on its own, because it designates a less specific dyad (hypertelorism and macrocephaly), and considers the old designation "polysyndactyly with peculiar skull shape" pejorative; Hootnick-Holmes syndrome has been proposed as the same entity [7]. GeneReviews maintains the recommendation not to use "Greig syndrome" as a synonym for GLI3-related GCPS [2].

The locus was mapped to 7p13 (now 7p14.1) through balanced translocations and overlapping deletions. In 1983, Tommerup and Nielsen described a family with t(3;7)(p21.1;p13) [38]. In 1991, Vortkamp, Gessler and Grzeschik studied three balanced translocations from different families and two sporadic cases with overlapping deletions at 7p13; two of the three translocations disrupted GLI3 in the first third of the coding sequence and the third had its breakpoint about 10 kb downstream of the 3′ end of the gene, which constituted the first association between GLI3 and GCPS [39]. In 1999, Kalff-Suske and colleagues described 15 novel point variants in 24 patients, distributed throughout the gene, and proposed haploinsufficiency as the mechanism [40].

1.2. Pallister-Hall syndrome

In 1980, Hall, Pallister and colleagues described six unrelated newborns with congenital hypothalamic hamartoblastoma, hypopituitarism, imperforate anus and postaxial polydactyly; some had a laryngeal cleft, abnormal lung lobulation, renal agenesis or dysplasia, short fourth metacarpals, nail dysplasia, multiple buccal frenula, micropenis, congenital heart disease and intrauterine growth restriction [41]. All had hypopituitarism and hypoadrenalism, all were sporadic cases of non-consanguineous parents, with an apparently normal karyotype, and all died in the neonatal period; five were born within an eight-month interval in different places, and several had prenatal exposure to insecticides or herbicides, with no exposure common to all six [41, 42]. The neuropathology was described in the same volume by Clarren and colleagues, in the second part of the work [42]. Iafolla and colleagues (1989) published a case with the delineation of "congenital hypothalamic hamartoblastoma syndrome (Pallister-Hall syndrome)"; Finnigan and colleagues (1991) extended it to other central nervous system malformations; Roscioli and colleagues (2005) described unreported skeletal findings; Narumi and colleagues (2010) reviewed the genital anomalies; and Li and colleagues (2015) described total colonic aganglionosis with imperforate anus in an infant with a severe form [42, 43]. Subsequent reports documented prolonged survival, familial cases and offspring of affected individuals, which dispelled the initial idea of a lethal syndrome; Hall later reviewed the early history of the syndrome and its evolution with modern medical genetics [42, 44, 45].

In 1996, an international workshop defined the clinical criteria for PHS and related phenotypes [46, 47]. In 1997, Kang and colleagues mapped the locus to 7p13 in a family with 22 affected members, with a trait of complete penetrance, variable expressivity and low morbidity (LOD score, logarithm of the odds of linkage, of 7.0 at θ = 0 in 60 individuals from two families), and in the same year identified GLI3 frameshift variants 3′ of the zinc-finger domains in two families, one of them de novo [48, 49]. PHS and GCPS came to be recognised as allelic disorders.

1.3. Later molecular milestones and reclassification

In 2005, the cohort of the US National Institutes of Health (NIH), with 135 individuals, showed that the type and position of the variant predict the phenotype: GCPS results from variants of all classes and PHS almost exclusively from truncating variants between nucleotides 1998 and 3481 of the reading frame [25]. In 2010, the same team extended the spectrum to patients who do not meet the classic criteria (phenotypes designated "sub-GCPS" and "sub-PHS"), to overlaps with oral-facial-digital syndrome and to non-syndromic polydactylies [24]. In 2008, Wallace and colleagues described somatic GLI3 variants restricted to the tissue of non-syndromic hypothalamic hamartomas with gelastic seizures [50]. Studies from 2015 to 2021 in French, Dutch and German cohorts refined the correlation between variant position and phenotype and proposed that isolated polysyndactyly is the mildest form of GCPS [51–53]. In 2017, Al-Qattan and colleagues grouped the nine conditions associated with GLI3 into four entities: the PPD-IV–GCPS–acrocallosal syndrome spectrum, postaxial polydactyly type A/B, PHS and the oral-facial-digital overlap [54].

1.4. Nomenclature

GeneReviews uses the designations "GLI3-related GCPS" and "GLI3-related PHS" [1, 2]. The PHS chapter was published on 25-05-2000, with comprehensive updates in 2003, 2005, 2008, 2012, 2014, 2017 and 18-08-2022 and a revision on 22-02-2024, which added c.2374C>T to the genotype–phenotype correlation; the GCPS chapter was published on 09-07-2001, with comprehensive updates in 2003, 2005, 2009, 2014, 2020 and 28-08-2025, the date of the current version [1, 2]. The designation "acrocephalopolysyndactyly" belongs to Noack, Carpenter and related syndromes (types I to IV) and should not be applied to GCPS or PHS, which are "cephalopolysyndactylies" [55]. For PHS, GeneReviews considers the term "hypothalamic hamartoblastoma syndrome" incorrect, because the suffix "blastoma" implies immature neural elements and because the term does not reflect the syndromic nature and is confused with isolated hamartomas; the descriptor CAVE (cerebroacrovisceral early lethality) is considered inappropriate, because few patients have early lethality; the term "sub-PHS" is considered of little use, and it is preferable to speak of mild PHS; and the acronym "HPS" should be reserved for Hermansky-Pudlak syndrome [1]. The cytoband "7p13", used in older sources, corresponds to the current 7p14.1 [23, 38].

1.5. Definition of the entities within scope

GCPS is defined by the association of preaxial polydactyly, with or without postaxial polydactyly, cutaneous syndactyly, macrocephaly and widely spaced eyes, with a heterozygous pathogenic GLI3 variant or a deletion at 7p14.1 that includes it [2]. GCPS due to contiguous gene deletion (GCPS-CGS) is the subtype in which the deletion includes neighbouring genes, with cognitive impairment, seizures and central nervous system anomalies being more frequent [7, 56]. PHS is defined, in clinical terms, by the presence of hypothalamic hamartoma and mesoaxial polydactyly in the proband, or, in clinical-molecular terms, by suggestive findings with a heterozygous pathogenic GLI3 variant [1, 24]. Postaxial polydactyly type A1/B (PAPA1/PAPB, OMIM 174200) and preaxial polydactyly type IV (PPD-IV, OMIM 174700) are apparently isolated digital phenotypes; the former is discussed as an entity in its own right or as mild PHS with subtle manifestations not sought, and the latter as GCPS without craniofacial manifestations [1, 2]. Non-syndromic hypothalamic hamartoma with a somatic GLI3 variant does not meet the criteria for PHS and can be considered a partial form [1].

2. Molecular Genetics and Pathophysiology

2.1. The GLI3 gene and protein

GLI3 (GLI family zinc finger 3; HGNC:4319; OMIM 165240; Ensembl ENSG00000106571; UniProt P10071) is located at 7p14.1 and spans about 300 kb of genomic DNA [7, 23]. The MANE Select reference transcript is NM_000168.6, which encodes the protein NP_000159.3 [2, 57]. Because of a cDNA sequencing error, older publications describe a longer reading frame, with a protein of 1596 amino acids; GeneReviews places the correction in GenBank record NM_000168.3, and variant numbering in older articles may therefore differ from the current numbering [1]. Direct comparison of the GenBank records showed, however, that NM_000168.2, used in the 2005 NIH cohort, already encodes the 1580-amino-acid protein and has a coding sequence of the same length as that of NM_000168.6 (4743 nucleotides, with no insertions or deletions, and two nucleotide differences, at c.547 and c.2993); the 2005 numbering therefore coincides with the current c. numbering, and the 1596-amino-acid annotation predates NM_000168.2 [25, 57]. The OMIM gene entry retains the 1596-amino-acid value of the original sequence and describes 15 exons, with exon 1 untranslated and the zinc fingers encoded by exons 10 to 13; OMIM associates four autosomal dominant phenotypes with GLI3 (GCPS, PHS, postaxial polydactyly types A1 and B and preaxial polydactyly type IV) [10]. The domains described in UniProt and cited by Siavrienė and colleagues are the repressor domain (amino acids 106–263), the five DNA-binding C2H2 zinc fingers (462–645), the proteolytic cleavage region (703–740) and the two transactivation domains, TA2 (1044–1322) and TA1 (1376–1580) [21, 40].

2.2. The hedgehog pathway and the activator–repressor balance

GLI3 is a zinc-finger transcription factor that acts downstream of Sonic hedgehog (SHH) in the SHH–PTCH1–SMO–GLI1/GLI2/GLI3 pathway; it is homologous to Drosophila cubitus interruptus and regulates genes such as HNF3β and those of the bone morphogenetic proteins [1]. In the absence of ligand, the full-length protein is proteolytically processed into a truncated repressor form (GLI3R), which is the predominant species; when the SMO (smoothened) receptor is activated, the full-length protein acts as an activator [23]. Processing depends on the primary cilium [38]. The balance between activator and repressor patterns the anteroposterior axis of the limbs, the craniofacies and the brain; digit separation occurs between days 52 and 55 after fertilisation [13, 23, 58]. In situ hybridisation in human embryos showed early expression of GLI3 in the target tissues of both syndromes [51].

2.3. Mechanism of Greig syndrome: haploinsufficiency

The most common mechanism of GCPS, and possibly the only one, is haploinsufficiency: whole-gene deletions produce a phenotype indistinguishable from that caused by nonsense or frameshift variants [2]. ClinGen classifies the gene–disease relationship as definitive and assigns GLI3 a haploinsufficiency score of 3 (sufficient evidence) and a triplosensitivity score of 0 [23]. GCPS results from variants of all classes: translocations, large deletions, exonic deletions and duplications, small in-frame deletions, missense, nonsense, frameshift and splicing variants [25]. Kalff-Suske and colleagues interpreted missense and splicing variants in the zinc fingers as causing functional haploinsufficiency (loss of DNA binding) and C-terminal variants as loss of transactivation [40]. The mechanism of 3′-end frameshift and nonsense variants and of missense variants has not been clarified [2]. In the 2005 NIH cohort, the missense variants p.C482W (at a conserved cysteine of a C2H2 zinc finger) and p.R625Q segregated with the disease and were absent from 160 control chromosomes, and p.R625W arose de novo, with confirmed paternity; exon-by-exon quantitative PCR identified a whole-gene deletion, deletions of exons 3–4 (with exon 2 spliced to exon 5 in the RNA) and of exon 2, and a duplication of exons 8–12 of at least 48.8 kb (notation of the article) [25].

2.4. Mechanism of Pallister-Hall syndrome: constitutive repressor

PHS-associated variants truncate the protein in the middle third, 3′ of the zinc fingers, without triggering nonsense-mediated mRNA decay (NMD); the resulting protein retains the repressor domain and the zinc fingers and loses the transactivation domains, behaving as a constitutive GLI3R [1, 25]. If the variant triggers NMD, the result is haploinsufficiency and GCPS [1]. Blake and colleagues showed, in an animal model, that constitutive expression of the repressor disrupts urogenital development in a cell-lineage-specific manner [22]. Saitsu and colleagues found that truncated GLI3 has greater repressor activity than the normal protein [59]. Hall described this effect as dominant negative [44]. Smith's Recognizable Patterns of Human Malformation uses, in the GCPS entry, the expression "activating mutations" for PHS, contradicting the description of a truncated repressor in the PHS entry of the same book; the constitutive repressor model is the one supported by GeneReviews and by functional studies [1, 38, 42].

2.5. Genotype–phenotype correlation by variant position

The mutational spectra of the two syndromes are largely distinct. The 2005 NIH cohort started from 153 individuals; after excluding 12 patients with GCPS due to a large deletion and 6 with PHS published separately, 135 remained (60 probands and 75 affected relatives: 20 probands with PHS and 26 relatives, 40 probands with GCPS and 49 relatives), with 47 pathogenic variants, and 38 of the 49 families with new clinical data were evaluated at the NIH [25]. Adding the published patients, GCPS resulted from 92 variants of all classes, and PHS (32 patients) only from nonsense, frameshift and one splicing variant; the dichotomy between truncating and non-truncating variants was associated with the phenotype with p<0.0001 (Fisher's exact test) [25]. Among frameshift, nonsense and splicing variants, those located 5′ of nucleotide 1998 or 3′ of nucleotide 3481 corresponded to 47 patients with GCPS and none with PHS (12 GCPS variants and no PHS variants 3′ of nucleotide 3481), and those located between these nucleotides to 8 patients with GCPS and 32 with PHS; deletions, translocations, small in-frame deletions, exonic deletions and duplications and missense variants corresponded to 37 patients with GCPS and none with PHS [25]. As the 2005 numbering (NM_000168.2) coincides with that of NM_000168.6, the limits of the PHS region are NM_000168.6:c.1998 to c.3481; at the protein level, the authors place the 5′ border at amino acid 666 and the C-terminal limit at amino acid 1161, and Démurger and colleagues describe the region as amino acids 667 to 1161 [25, 51, 57]. The transition at nucleotide 1998 relates to the zinc-finger domain and the proteolytic processing site, and that at nucleotide 3481 to the transactivation domain [24]. Within each region there is no correlation between position and severity [1, 2]. The only PHS splicing variant in the 2005 cohort, in intron 14 of family PH02 (a large family with mild PHS and 22 affected members), produced three transcripts, two with a frameshift and one with an in-frame deletion of 75 amino acids, and was interpreted as an allele generating only part of the repressor transcripts, consistent with the mild phenotype [25].

Table 2.1 — Correlation between variant class and position and phenotype

The most studied exception is the variant NM_000168.6:c.2374C>T p.(Arg792Ter), located in the middle of the PHS region and associated with GCPS. In 2005, six independent families had been described (three from the NIH cohort and three from Kalff-Suske 1999 and Debeer 2003); the 2010 series added the eighth family, all with GCPS, and Abdullah and colleagues described it in a Pakistani family, which supports the "at least nine apparently unrelated families" of the 2025 update of the GCPS GeneReviews; the 2024 revision of the PHS GeneReviews and the legend of figure 1 of the GCPS GeneReviews retain the count of six [1, 2, 24, 25, 62]. As to mechanism, Johnston and colleagues observed similar levels of mutant and normal mRNA in a lymphoblastoid line from family G44 (data not shown in the article), whereas the 2010 series, citing Furniss 2007, and OMIM describe nonsense-mediated mRNA decay, which would explain haploinsufficiency [10, 24, 25]. In ClinVar (VCV 13828) the variant is classified as pathogenic by several submitters, with the conditions PAPA1, PHS and GCPS [63]. In 2010, the total of published variants in typical GCPS and PHS was 147: in GCPS, 31 large deletions or duplications, 5 translocations, 9 missense, 1 in-frame deletion, 11 splicing and 54 frameshift or nonsense; in PHS, 1 splicing and 35 frameshift or nonsense (p<0.0001); the frameshift and nonsense variants of typical GCPS were distributed as 31 5′, 9 in the PHS region and 14 3′ [24].

Later studies added three refinements. First, Baas and colleagues reviewed the 2019 Human Gene Mutation Database (HGMD) Professional and the patients of the Sophia Children's Hospital (Rotterdam): 297 patients (251 published and 46 local) from 155 families, with 127 different variants, of which 32 were large deletions, copy-number variants or translocations; they excluded PHS, acrocallosal syndrome, isolated preaxial or postaxial polydactylies and, in the genotype analysis, structural variants [52]. Latent class analysis identified a posterior class (n=88; postaxial polydactyly of the hand in 96.6% and of the foot in 69.3%) and an anterior class (n=201; preaxial polydactyly of the foot in 95.5%, syndactyly of the foot in 83.1% and preaxial polydactyly of the hand in 52.7%); missense (54 versus 6) and splicing (23 versus 1) variants were concentrated in the anterior class (p<0.001) [52]. Frameshift and nonsense variants in the activator domain were associated with postaxial polydactyly of the hand (odds ratio, OR, 12.7; 95% CI 5.2–31.0) and of the foot (OR 33.9; 15.1–76.0) and, conversely, with a lower frequency of preaxial polydactyly of the foot (OR 0.1; 0.032–0.14) and of syndactyly of the foot (OR 0.1; 0.054–0.19); agenesis of the corpus callosum occurred in 11 of 41 patients in the posterior class versus 3 of 75 in the anterior class (OR 8.8; p<0.001), and 11 of the 14 patients with agenesis had a truncating variant in the activator domain [52]. The text of the article reverses, in one paragraph, the distribution of missense variants between the classes; table 3, the abstract and the conclusion agree with one another and are the version used here [52]. The authors recommend distinguishing preaxially and postaxially oriented phenotypes to estimate the risk of agenesis of the corpus callosum, because brain imaging is rarely reported in the series [52]. Second, Sczakiel and colleagues (48 new cases from 16 families and a review of 314 published variants) found no difference in location between GCPS variants and those of isolated polysyndactyly, confirmed PHS in the central third and proposed a binary division between PHS and GCPS/isolated polysyndactyly, contrary to the old attribution of isolated polysyndactyly to the C-terminal third [53]. Third, in the French cohort of Démurger and colleagues (76 individuals from 55 families: 55 with GCPS from 38 families, including 6 with complete gene deletion, and 21 with PHS from 17 families, including 7 fetuses; sequencing of the 14 coding exons on NM_000168.5, with fluorescence in situ hybridisation (FISH), multiplex ligation-dependent probe amplification (MLPA) or array comparative genomic hybridisation when negative), GCPS probands had nonsense variants in 8 of 38 (21%), frameshift in 17 (45%), missense in 6 (16%), all in the DNA-binding domain (amino acids 462–645, exons 10–13), splicing in 1 (3%) and whole-gene deletion in 6 (16%); the 17 PHS variants were all truncating (12 frameshift and 5 nonsense), 3′ of the DNA-binding domain and within the PHS region, and those of fetuses with severe forms clustered in the middle third, between c.2941 and c.3324, in the transactivation and CREB-binding protein (CBP) binding region [51]. Variant position correlated with corpus callosum anomalies in GCPS (7 of 9 with a truncating variant in the final third) [51]. In the systematic review of 78 patients with PHS by McClelland and colleagues, patients with congenital anomalies of the kidney and urinary tract (CAKUT) had mainly nonsense substitution variants in the middle third, and patients without CAKUT mainly deletions causing a frameshift [64].

2.6. Structural variation and contiguous gene syndrome

Deletions that include GLI3 range from 1 nucleotide to almost 1 Mb in the typical forms, and insertions from 1 nucleotide to about 60 kb [7]. In the 2003 NIH series, fluorescence in situ hybridisation and polymorphic marker analysis detected deletions in 11 of 34 patients with GCPS features; 9 of the 11 had intellectual disability or developmental delay, and the six characterised deletions measured between 151 kb and 10.6 Mb, with no common sequences at the breakpoints [56]. GeneReviews associates deletions of more than 300 kb with a higher frequency of intellectual disability, seizures and central nervous system anomalies; this threshold is from GeneReviews and does not appear in the abstract of Johnston 2007, which describes a targeted comparative genomic hybridisation microarray (about 14,000 probes over 10 Mb, one probe per 730 bp) that confirmed 15 of 16 previous results (14 deletions and 1 duplication), showed that the 16th result by quantitative polymerase chain reaction (PCR) was a false positive and sizes the deletion better than FISH [2, 65]. Kroisel and colleagues described five patients with GCPS and a 7p13 microdeletion (including de novo monozygotic twins) with moderate psychomotor delay, seizures, muscle fibre anomalies, heart disease, hyperglycaemia and hirsutism, and recommended looking for the microdeletion in all atypical GCPS [66]. The 2008 review proposes as a rule of thumb that alterations smaller than about 1 Mb cause typical GCPS and those larger than about 1 Mb cause GCPS-CGS, with normal or small head circumference and a hypoplastic corpus callosum [7]. The two thresholds come from different series and describe a risk gradient, not a validated cut-off.

The deletion may include GCK, whose loss of one copy causes MODY2-type diabetes (maturity-onset diabetes of the young type 2), described in patients with GCPS-CGS [26, 67]. Schwarzbraun and colleagues, in five patients with a 7p deletion (including monozygotic twins), also identified loss of PGAM2, relevant because of the risk of recessive myopathy if the other allele carries a variant [26]. Translocations and inversions disrupting GLI3 are rare. GeneReviews cites the family with t(3;7)(p21.1;p13) of Tommerup and Nielsen (1983), a large family with a reciprocal translocation t(6;7)(q27;p13) (Krüger 1989) and a boy with a 14 Mb deletion at 7p13-p14 resulting from a balanced paternal (5;7) insertion (Schulz 2008) [2]. In the review of 43 skeletal genes in the 100,000 Genomes Project (Genomics England), Pagnamenta and colleagues identified inversions of 1.2 and 14.8 Mb disrupting GLI3 in two families with GCPS, one of them with a mixed phenotype due to a breakpoint about 45 kb from HOXA13, after diagnostic odysseys of 9 to 20 years; these were the first large inversions described in GLI3 and show that balanced structural variants escape copy-number methods [2, 60]. Kozma and colleagues compiled the published deletions larger than 1 Mb, of which only six had exact genomic coordinates: 6.2 Mb (Niida 2015), 7 and 9 Mb (Démurger 2015), 6.0, 6.8 and 8.3 Mb (Hurst 2011), 14 Mb (Schulz 2008), 1.8 to 10.3 Mb in nine patients (Johnston 2003 and 2007) and 4.5 to 15 Mb (Kroisel 2001 and Schwarzbraun 2005), to which they added one of 18.37 Mb (section 10.3) [9].

2.7. Allelic disorders and rare phenotypes

Table 2.2 — GLI3 allelic disorders (adapted from GeneReviews, Table 3, and primary sources)

In the oral-facial-digital overlap, the 2010 NIH series included patients with truncating variants such as p.K693X, p.Q993X and p.G1001AfsX2 (historical notation of the article, not normalised), some of whom met the criteria for PHS [24]. Ito and colleagues described, in Japan, an overlapping GCPS/PHS phenotype with agenesis of the gallbladder and pancreas associated with the missense variant p.(Pro719Ser) in the proteolytic cleavage region, classified by the authors as likely pathogenic [61]. Turner and colleagues described, in a sporadic case of PHS, a de novo 72 bp insertion of mitochondrial DNA (sequence overlapping the tRNA genes MT-TS2 and MT-TL2) in exon 14, on the paternal allele, which creates a premature termination codon; the patient had a hypothalamic hamartoma without endocrine changes or seizures, scars from excision of supernumerary ulnar digits, metacarpal fusion and a bifid epiglottis [10].

2.8. Somatic mosaicism in hypothalamic hamartoma

About 5% of hypothalamic hamartomas are associated with PHS [20]. Wallace and colleagues showed in 2008 that somatic GLI3 variants restricted to the hamartoma can cause hypothalamic hamartoma and gelastic seizures [50]. In the same year, Craig and colleagues compared DNA from blood and resected tissue in 55 patients with sporadic hamartoma and intractable epilepsy: they found no causal germline variants in GLI3, but identified chromosomal anomalies involving the GLI3 locus (a 7p anomaly and intragenic loss of heterozygosity) in 8 of the 55 tissues (15%) [20]. Saitsu and colleagues studied 18 blood–hamartoma pairs by exome sequencing: they found somatic truncating variants in GLI3 in two patients and in OFD1 in three (all male), with allele fractions between 7% and 54%; nine had no somatic variant identified [59]. Green and colleagues, in 27 patients with resected tissue, identified biallelic variants (one germline and one somatic) in DYNC2I1, DYNC2H1, IFT140 or SMO in seven, and a single somatic variant in GLI3 or OFD1 in eight, which explained 15 of 27 cases (56%) and led to the reconceptualisation of sporadic hamartoma as a ciliopathy [75]. In the Paris series of Sami and colleagues, seven of nine hamartomas had pathogenic somatic variants in GLI3, OFD1 or PRKACA [76]. A 2023 review notes that most non-syndromic hamartomas have no family history and that, in PHS, the GLI3 variant is germline, inherited or de novo, in most patients [77]. The 2026 review describes a continuum between somatic mosaicism and germline variants in ciliary genes [78]. GeneReviews notes that the pathogenicity of some of the somatic variants described is questioned, because they violate the known mechanism of GLI3 pathogenesis [1].

2.9. Penetrance, expressivity and germline mosaicism

In PHS there are no published cases of incomplete penetrance, and variability within the same family appears to be low [1]. Ng and colleagues described two siblings with severe PHS (intellectual disability, behavioural changes and refractory seizures) whose parents were clinically normal on examination, laryngoscopy and brain MRI; the variant was not detected in the parents' blood, and intragenic markers showed the same maternal allele in both children, with confirmed paternity, which established germline mosaicism [79]. In GCPS, cases of apparent non-penetrance and of gonadal or somatic-gonadal mosaicism have been described in rare families; when the variant arose in the parent, the parent may have somatic and gonadal mosaicism and be minimally or mildly affected, whereas the offspring, with the constitutional alteration, may have a more severe phenotype [2, 51]. In the French series, the c.1543_1544dup duplication of two affected siblings was present at a low level in the blood of the asymptomatic father (family G068), a GLI3 deletion was detected by FISH in only 56% of the blood cells of a patient with bilateral preaxial polydactyly of the feet and delay (G059), and an apparently sporadic case inherited the variant from a healthy mother without signs of GCPS (G070) [51]. Debeer and colleagues described a four-generation family with the missense variant p.R625W and a carrier without clinical signs, the same family recorded in OMIM as evidence of incomplete penetrance [10, 80]. GLI3 germline mosaicism is rare: by 2015, the two siblings with PHS of Ng and colleagues and family G068 had been described [51, 79]. The sources diverge on variability: Smith's describes great variability between and within families, whereas GeneReviews and the 2008 review consider that variability within a family is lower than between families, so that familial severity is the best predictor [2, 7, 38].

2.10. Variants cited in this article, normalised

The variants in table 2.3 were normalised in VariantValidator (GRCh38, NM_000168.6) and checked in ClinVar on 17 October 2026 [57, 63]. Historical notations were not converted manually.

Table 2.3 — GLI3 variants cited, with normalised notation

Tao and colleagues described in 2025 ten variants in ten Chinese families with GCPS, PHS, PAP-A/B and PPD-IV, seven of them not recorded in HGMD, among which c.2374C>T p.(Arg792*); the abstract does not state the transcript [85]. Abdullah and colleagues described, in four Pakistani families with GCPS, the variants p.(Tyr146Leufs*19), p.(Glu99Serfs*60), p.(Thr541Arg) and the recurrent p.(Arg792*) [62]. The variant p.(Pro668Leu), at the first codon of the PHS region, has conflicting classifications in ClinVar and was described with isolated postaxial polydactyly, which illustrates the difficulty of interpreting missense variants near the border of the region [63, 83].

The list of 22 OMIM allelic variants coincides, for the variants in table 2.3 that it includes, with the c. notation of NM_000168.6: p.(Arg643Ter) corresponds to allele .0008 (1927C>T, in a family with postaxial polydactyly A1 and B), p.(Ala934Pro) to allele .0013 (2800G>C, child with agenesis of the corpus callosum and severe disability, classified as severe GCPS and not as acrocallosal syndrome because of a heterozygous GLI3 variant) and p.(Arg792Ter) to allele .0016 (2374C>T, with demonstrated mRNA decay) [10]. p.(Gly727Arg), described in 1999 as a cause of postaxial polydactyly A/B, was reclassified in OMIM in 2018 as a variant of uncertain significance, because it is present in 693 of 121,412 alleles in the ExAC database, with 4 homozygotes (allele frequency 0.005); the absence of segregation in the family of Siavrienė and colleagues is consistent with this reclassification [10, 21]. The same entry records p.(Arg290Ter) (868C>T) in four unrelated families with GCPS and in a patient with preaxial polydactyly type IV whose carrier father had only preaxial polydactyly of the feet, a case that fuelled the discussion of PPD-IV as minimal GCPS, and p.(Arg625Trp) in a four-generation family with GCPS and transmission through a carrier without clinical signs, evidence of incomplete penetrance in GCPS [10].

2.11. Differences between entities and modifiers

The sources consulted do not describe validated genetic modifiers of expressivity. The phenotype depends mainly on the variant class (haploinsufficiency versus constitutive repressor), position (thirds of the gene and activator domain), deletion size (contiguous genes) and somatic or germline status [7, 25, 52, 75]. The clinical overlap between GLI3 PHS and Pallister-Hall-like syndrome, recessively inherited through biallelic SMO variants (OMIM 241800), shows that the PHS phenotype results from disruption of the pathway and not of a single protein [20, 86, 87].

3. Clinical Spectrum and Manifestations by System

3.1. Ascertainment bias and interpretation of frequencies

The published frequencies for both syndromes are subject to ascertainment bias. In PHS, most patients were ascertained because they had hypothalamic hamartoma and polydactyly, so stating that these findings are common is circular reasoning; GeneReviews therefore presents frequencies semiquantitatively and considers that only large exome or genome studies will resolve the question [1]. In GCPS, the GeneReviews percentages are based on about 300 published individuals with a pathogenic variant, including cases of isolated preaxial polydactyly, and the Smith's percentages come from clinical series with no stated denominator [2, 38]. The assumption that PHS is severe and GCPS mild is incorrect: a minority of patients with PHS have severe multiple anomalies, and most have polydactyly, asymptomatic bifid epiglottis and hypothalamic hamartoma, and are often diagnosed as postaxial polydactyly type A if not carefully evaluated [1].

3.2. Pallister-Hall syndrome

Table 3.1 — Frequency of manifestations in PHS (GeneReviews, semiquantitative estimate) and cohort data

Table 3.2 — Manifestations of PHS in two cohorts with a denominator

Sources: [25, 51]. Both cohorts are referral cohorts; in the NIH cohort, the inclusion criteria required hamartoma and insertional polydactyly in the proband, and the large family PH02 is excluded from the individual data. Insertional polydactyly is defined by a floating metacarpal or metatarsal between the 2nd and 5th, or by fusion of metacarpals or metatarsals other than the first or the most posterior [25].

3.2.1. Hypothalamic hamartoma

The hypothalamic hamartoma is a malformation and not a tumour: it is a mass in the floor of the third ventricle, posterior to the optic chiasm, which grows at the rate of the surrounding brain tissue or more slowly and can reach 4 cm in its greatest dimension [1]. In the original series, the hamartoblastoma occupied the inferior surface of the brain, from the optic chiasm to the interpeduncular fossa, replacing the hypothalamus and other nuclei originating from the embryonic hypothalamic plate, with pituitary aplasia or dysplasia [42]. There is little correlation between size and the presence or severity of symptoms, and most hamartomas in PHS are asymptomatic [1]. No patient with PHS has shown visual field loss, even with a hamartoma adjacent to the chiasm [1]. In the NIH series of Boudreau and colleagues, MRI confirmed a hamartoma in 37 of 40 patients with PHS; the PHS hamartoma was mostly isointense to grey matter on T2, whereas all but one of the isolated hamartomas were hyperintense [90]. In a series of 18 patients with hamartoma (mostly non-syndromic), the hamartomas were isointense on T1, isointense on T2 in 88.9% and hyperintense in 11.1%, and half had other brain anomalies (focal cortical dysplasia, amygdala enlargement, hypoplasia of the corpus callosum) [91].

3.2.2. Endocrine manifestations

The endocrine spectrum ranges from isolated growth hormone (GH) deficiency or isolated precocious puberty to potentially fatal panhypopituitarism [1]. Cortisol deficiency can occur in non-familial PHS, but appears to be rare in familial PHS [1]. In the NIH series, central precocious puberty was more likely in isolated hamartoma than in PHS, in which other endocrine disturbances predominated [90]; in the French series, four of 13 patients assessed had isolated GH deficiency and one had panhypopituitarism, and four of five fetuses had adrenal hypoplasia [51]. The main cause of neonatal death is hypoadrenalism; many long-term survivors required levothyroxine, GH and corticosteroids from early on and glucose infusions in the neonatal period [42]. Some forms feature micropenis and testicular hypoplasia [41, 42].

3.2.3. Neurological and behavioural manifestations

Hamartoma-associated epilepsy manifests as gelastic seizures, focal seizures with clonic movements of the chest and diaphragm that mimic laughter, and other seizure types [1]. Boudreau and colleagues compared, at the NIH, 40 patients with PHS and 16 with isolated hamartoma: all 16 with isolated hamartoma had seizures, versus 13 of the 40 with PHS (all of these with hamartoma); in isolated hamartoma, seizures began earlier, were more frequent and harder to control and were accompanied by more behavioural and developmental problems, whereas in PHS seizures were generally well controlled with antiseizure medication [1, 90]. The authors recommend looking for signs of PHS in every patient with hamartoma, especially before considering surgery [90]. The 2021 international review explains the difference by location: the parahypothalamic hamartoma, frequent in syndromic forms such as PHS, mainly causes central precocious puberty without severe neurological comorbidity, whereas the intrahypothalamic hamartoma causes drug-resistant epilepsy, neurocognitive and behavioural sequelae and endocrine dysfunction [17]. This generalisation has exceptions: there are patients with PHS and severe epilepsy, such as the two siblings with germline mosaicism described by Ng and colleagues and the Colombian adult with refractory gelastic seizures and an intelligence quotient (IQ) of 78 [79, 82]. Some patients have behavioural manifestations and few have severe intellectual disability with behavioural disorders [1]. In the prospective neuropsychiatric assessment of 19 patients with PHS (7 to 75 years), 7 of 14 adults had at least one DSM-IV Axis I diagnosis and another 3 had delay or neuropsychological deficits, with no characteristic psychiatric phenotype and frequencies not differing from those expected; comparison with the general population was not possible because of ascertainment bias [89]. In the 2005 NIH cohort, 13 of 23 patients assessed had developmental delay or intellectual disability and 17 of 24 had seizures and/or headaches (recorded in the same column), and in the French series intellectual efficiency was normal in most, with mild delay in 3 and gelastic epilepsy in 2 [25, 51]. GeneReviews recommends, from 12 months of age, screening for sleep disorders, attention-deficit/hyperactivity disorder (ADHD), anxiety and autism spectrum disorder (ASD) [1]. The Oxford Desk Reference includes GLI3 among the genes associated with Lennox-Gastaut syndrome [92]. Holoprosencephaly with median cleft lip and palate, arhinencephaly, Dandy-Walker malformation, polymicrogyria and occipital encephalocele have been described as occasional findings [42, 93].

3.2.4. Airway

Bifid epiglottis is an anteroposterior midline cleft involving at least two-thirds of the epiglottic leaf; it is almost always asymptomatic, very rare in other syndromes and rare as an isolated malformation, which makes it useful for diagnosis [1]. Outside PHS, bifid epiglottis has been described in congenital hypothyroidism due to FOXE1 and, less often, in short-rib dysplasia type III, and laryngeal anomalies are part of hydrolethalus syndrome and oral-facial-digital syndrome type VIII (section 6) [94–97]. In the NIH laryngoscopy study, 15 of 26 patients with PHS (58%) had a bifid or cleft epiglottis and none of the 14 patients with GCPS; the malformed epiglottis was asymptomatic in all patients assessed prospectively, and posterior laryngeal clefts were uncommon and restricted to severely affected patients [88]. In the 2005 cohort, bifid epiglottis was present in the 19 patients examined, and in the French series in half of those examined, always asymptomatic in postnatal cases; the denominators differ because not all patients were examined [25, 51]. More extensive laryngotracheal clefts cause severe symptoms and posterior ones can be fatal [1]. Cricoid cartilage stenosis with anterior synechia of the vocal cords, which required tracheostomy at 13 months, and subglottic stenosis have been described [98, 99]. Smith's adds tracheal cartilage dysplasia, tracheal cleft, absent lung and abnormal lung lobation [42].

3.2.5. Limbs

Mesoaxial (central or insertional) polydactyly consists of six or more well-formed digits with a Y-shaped metacarpal or metatarsal and is the most specific digital finding [1]. Postaxial polydactyly type A (well-formed digit on the ulnar or fibular side) and type B (rudimentary digit or nodule) is probably more common than mesoaxial polydactyly, but is non-specific [1]. In the French series, the Y-shaped metacarpal or metatarsal was present in 83% of patients with PHS and, in the remainder, there were numerical or morphological anomalies of the metacarpals or metatarsals; PHS was confirmed molecularly in four individuals without polydactyly, and the authors consider the Y-shaped metacarpal a more significant criterion than insertional polydactyly, because central polydactyly and Y-shaped metacarpals are very rare in other syndromes [51]. Smith's further describes nail dysplasia, variable syndactyly, oligodactyly, a small and distal fourth metacarpal with one or two associated small fingers, a dysplastic fourth metatarsal and distal shortening of the limbs, particularly the arms [42]. In the French series, four fetuses with PHS had mesomelia or micromelia (three with oligodactyly, clubfoot and arthrogryposis) and one had agnathia, which extends the spectrum to lethal prenatal forms, including acromesomelic shortening with radioulnar bowing and hypoplasia or agenesis of the tibia and fibula [51].

3.2.6. Digestive tract, kidney and genitalia

Imperforate anus, rectal atresia and colonic aganglionosis are part of the spectrum [42]. In the systematic review by McClelland and colleagues (78 published patients with PHS defined by hamartoma, mesoaxial polydactyly and a truncating variant in the middle third), CAKUT was present in 21 (26.9%), mainly renal hypoplasia (with or without dysplasia) and renal agenesis, with equal frequency bilaterally and unilaterally; CAKUT was associated with craniofacial defects, bifid epiglottis and differences of sex development of the external genitalia [64]. Narumi and colleagues reviewed 12 published patients with a GLI3 variant and genital anomalies: males had hypospadias, micropenis and a bifid or hypoplastic scrotum, and all females had hydrometrocolpos and/or vaginal atresia; the authors attributed these anomalies to SHH dysregulation and not to hormonal changes, and recommend investigating them even in the absence of hypopituitarism [43]. In the French series (table 3.2), half of the patients had an imperforate or anteriorly placed anus (all fetuses), 41% renal anomalies and half genitourinary anomalies [51]. A Japanese patient with an overlapping phenotype had agenesis of the gallbladder and pancreas [61].

3.2.7. Other systems and occasional findings

Smith's lists heart defects (atrioventricular septal defect, patent ductus arteriosus, ventricular septal defect, mitral and aortic valve disease, coarctation of the aorta), craniofacial anomalies (flat midface with a midline capillary haemangioma, short nose, anteverted nares, auricular anomalies, micrognathia), multiple oral frenula and, as occasional findings, clefts, microphthalmia, coloboma, natal teeth, vertebral and rib anomalies, hip dislocation, and pancreatic and thyroid hypoplasia [42]. An adult with PHS had hyperinsulinaemic hypoglycaemia, but the association with the syndrome has not been established [81].

3.2.8. Severity and age at diagnosis

The spectrum ranges from the sporadic neonatal lethal form of the original series to the mild familial form, with complete penetrance and low morbidity, of the family with 22 affected members [41, 48]. De novo cases are generally more severe than familial ones [1]. The diagnosis may be made only in adulthood: the Saudi patient in section 10 was diagnosed at 26 years of age, and the Colombian adult at 23 years, years after the identification of the hamartoma [81, 82].

3.3. Greig cephalopolysyndactyly syndrome

Table 3.3 — Frequency of manifestations in GCPS according to the sources

Sources: [2, 38, 100]. The differences between columns reflect different populations and criteria (molecularly confirmed cases versus older clinical series) and should not be combined.

Table 3.4 — Manifestations of GCPS in three cohorts with a denominator

Sources: [25, 51, 52]. These are referral cohorts with different criteria: in the NIH cohort, preaxial polydactyly was a selection criterion (hence 100%); the Baas cohort includes published cases and excludes structural variants from the genotype analysis. The denominators vary because the findings were not reported in all patients.

3.3.1. Skull and face

Macrocephaly is defined as head circumference above the 97th centile for age and sex and should be interpreted with caution when a parent has benign familial macrocephaly (OMIM 153470); some patients have a high anterior hairline and a prominent or bossed forehead [2]. Widely spaced eyes are defined as an interpupillary distance more than 2 standard deviations above the mean in newborns of 27 to 41 weeks, above the 97th centile between 0 and 15 years, or subjectively increased; telecanthus is possible but less distinctive, and is associated with a broad nasal root [2]. The 2008 review points out pitfalls: mild hypertelorism is seen as an attractive feature, familial macrocephaly is common and there are families with several affected members almost without craniofacial signs; the macrocephaly of typical GCPS is not accompanied by a central nervous system anomaly [7]. The facial features may be subtle, especially in infancy [68]. Metopic craniosynostosis, sometimes with sagittal synostosis, has been described in five children with trigonocephaly, preaxial and postaxial polydactyly and cutaneous syndactyly, three of whom had a complete GLI3 deletion and a previous diagnosis of Carpenter syndrome [101]. The atlas of skeletal dysplasias states that GCPS and PHS do not have craniosynostosis, which contradicts these cases; craniosynostosis is rare in GCPS, but it does occur [7, 55, 101]. In the French cohort of Démurger and colleagues (55 patients with GCPS: 49 with a point variant and 6 with a large deletion), only 10 probands met all four clinical criteria (preaxial polydactyly, cutaneous syndactyly, widely spaced eyes and macrocephaly) and 17 had absent or very subtle craniofacial features; at least 13% had macrosomia, 2 had craniosynostosis, and 9 patients in the cohort had corpus callosum anomalies, 7 of whom had a truncating variant in the C-terminal domain [5, 51].

3.3.2. Limbs

At least one limb has preaxial polydactyly (total or partial duplication of the first ray), a markedly broad hallux or a markedly broad thumb (increased width without increased dorsoventral dimension); the other limbs may have preaxial or postaxial polydactyly or five normal digits, and the postaxial polydactyly may be type A, B or intermediate [2]. Broadening of the first digit may be only radiographic, which GeneReviews considers difficult to assess when diagnosing a proband [2]. In the NIH experience, the most frequent finding is postaxial polydactyly of the hands with preaxial polydactyly of the feet, and severity varies between individuals and between limbs of the same individual, from an apparently normal limb or subtle broadening to higher-order polydactyly (one patient with octodactyly) [7]. In the Leuven series (13 patients), the hands never had preaxial polydactyly, but rather postaxial polydactyly, broad thumbs, clinodactyly of the thumb and variable syndactyly; all had syndactyly of the toes and an anomaly of the hallux, with syndactyly most frequent between the 1st, 2nd and 3rd toes [102]. Cutaneous syndactyly ranges from absent to complete involvement of all digits, with an appearance similar to that of Apert syndrome in extreme cases, and occasionally there is fusion of parts of the distal phalanges [2, 7]. Bone age may be advanced [38].

3.3.3. Central nervous system and development

Developmental delay, intellectual disability and seizures are uncommon (less than about 10%) and more likely with central nervous system malformations, hydrocephalus or deletions of more than 300 kb, according to GeneReviews [2]. In the French series, 18 patients underwent brain MRI: 7 had ventricular dilatation, 9 corpus callosum anomalies (7 with a truncating variant in the final third and only 1 with a large deletion) and 2 cerebellar hypoplasia without the molar tooth sign; 5 of 6 patients with a large deletion had developmental delay, and 9 without a deletion had mild delay, mainly in fine motor skills (7 with a C-terminal variant) [51]. Hypoplasia or agenesis of the corpus callosum is reported in about 20% in the GeneReviews summary, whereas the differential diagnosis table of the same document considers it rare in GCPS; in the review by Baas and colleagues it was reported in 16 of 145 patients with described imaging and was associated with truncating variants in the activator domain [2, 52]. Agenesis of the corpus callosum, with a prevalence of 3 to 7 per 1000 in the population and of up to 2% in children with developmental disorders, is a poorly specific finding and is sometimes incidental with normal development [103]. The 2008 review estimates that the risk of intellectual disability in typical GCPS has not been shown to exceed the population baseline risk (2–3%) and that it is concentrated in GCPS-CGS [7]. In the Leuven series, three of 13 patients had intellectual disability, associated with a large deletion, without structural brain anomalies [102]. Isolated cases of tumours (leukaemia, gliomas) have been described, with no attributable relative risk [7].

3.3.4. Other findings

Short or downslanting palpebral fissures, camptodactyly, heart defects, hyperglycaemia, hirsutism, inguinal, umbilical and diaphragmatic hernias, cryptorchidism and hypospadias are occasional [7, 38]. Macrosomia, present in at least 13% of the French series, raises the differential with Simpson-Golabi-Behmel syndrome type 1 [51]. In the Lithuanian family in section 10 there was progressive macrocephaly up to 61 cm in the adult and mild hydrocephalus detected by computed tomography (CT) at 27 years [21].

3.4. GCPS due to contiguous gene deletion

GCPS-CGS is a clinical entity distinct from typical GCPS, with frequent cognitive impairment and seizures, a hypoplastic corpus callosum and normal or small head circumference [7]. Children with a deletion have additional developmental delay and often lack macrocephaly [68]. The overlap with acrocallosal syndrome justifies molecular distinction [56]. When the deletion includes GCK, there is MODY2-type diabetes [26, 67]. In the 2010 NIH series, three deletions of 4.1 to 12.2 Mb were associated with microcephaly or normocephaly, cognitive impairment and seizures [24]; in the French series, two patients with deletions of 7 and 9 Mb had preaxial polydactyly of the feet, developmental delay, strabismus and cataract [51]. Kroisel and colleagues described the phenotype of five patients with a 7p13 microdeletion (section 2.6) [66]. Kozma and colleagues described an 18.37 Mb deletion with 205 genes, including 22 OMIM genes with a phenotype, among them GCK, ANLN, POU6F2, CCM2, IKZF1, CDK13, CAMK2B and RALA, which requires surveillance for the corresponding comorbidities (case in section 10.3) [9].

3.5. Apparently isolated polydactylies due to GLI3

Isolated PAP-A is a well-formed supernumerary postaxial digit; its relationship with PHS is debated, because mild forms of PHS with asymptomatic bifid epiglottis, hamartoma or anal stenosis pass for isolated polydactyly if they are not sought [1]. In the 2010 NIH series, only one of five patients with isolated PAP-A had a GLI3 variant (c.874C>T p.R292C, upstream of the zinc fingers), and in the "sub-PHS" group a female proband with bilateral mesoaxial polydactyly, isolated GH deficiency without hamartoma and bifid epiglottis had the truncating variant c.3887_3894del p.L1297SfsX4, outside the PHS region, shared by three relatives with postaxial polydactyly of two to four limbs and bifid epiglottis (notation of the article) [24]. In four consanguineous Pakistani families with isolated postaxial polydactyly, variants in GLI1, IQCE and GLI3 (p.(Pro668Leu) and p.(Ala1522ProfsTer2)) were identified [83]. In a Chinese cohort of 44 fetuses with polydactyly studied by copy-number sequencing and trio exome sequencing, the overall yield was 36.4% (6 aneuploidies, 1 copy-number variant and 9 monogenic diagnoses, one of them in GLI3): 9.5% in isolated polydactyly versus 60.9% in polydactyly associated with other anomalies, 64.7% in bilateral versus 18.5% in unilateral, and 65.0% in postaxial versus 12.5% in preaxial; all fetuses with isolated polydactyly and 34.8% of the remainder were born alive [104]. PPD-IV corresponds to polysyndactyly with broad or bifid thumbs and halluces, syndactyly and sometimes postaxial polydactyly, with greater involvement of the feet, and is generally considered part of the GCPS spectrum [21, 53, 68]. Wang and colleagues described an infant with Leber congenital amaurosis type 10 due to CEP290 variants and postaxial polydactyly type A1 due to a GLI3 variant inherited from the father, initially diagnosed as Bardet-Biedl syndrome; the dual aetiology was identified by exome sequencing only after six years [105].

3.6. Hypothalamic hamartoma of all causes

The following data come from mostly non-syndromic series and do not translate directly to PHS. The International League Against Epilepsy (ILAE) classification of electroclinical syndromes includes gelastic seizures with hypothalamic hamartoma among the distinctive constellations [106]. Gelastic seizures (uncontrollable mirthless laughter) typically begin before 1 year of age and are often not recognised in the first years; between 2 and 7 years they may evolve into focal seizures with impaired awareness, dacrystic seizures, atypical absences, tonic, atonic and generalised tonic-clonic seizures and infantile spasms [17]. Drug resistance is reported in 50% to 100% of series, with selection bias [17]. Psychiatric comorbidity affects more than half of children, with rage attacks and aggression in 50% to 80%, more common in boys, with intellectual disability, early seizure onset and multiple seizure types; cognitive impairment exceeds 80% and is progressive in half, which justifies the designation of epileptic encephalopathy [17]. The typical preoperative endocrinopathy is central precocious puberty, due to premature pulsatile release of GnRH [17]. The 2026 international consensus summarises similar data: gelastic seizures at onset in up to 77% of patients; psychiatric comorbidity in 55% (ADHD, oppositional defiant disorder, conduct disorder, ASD and affective disorders, at rates similar to those of other lesional epilepsies), with the exception of rage attacks, characteristic of hamartoma, and behavioural comorbidities that may precede seizures; presurgical delay or intellectual disability in about 50% (the 2021 review indicated cognitive impairment in more than 80%; both estimates come from series with selection bias), with deficits in visual and verbal memory, executive function and processing speed, worse with early onset, frequent seizures, larger lesions and more antiseizure medications; and central precocious puberty in 40% to 67%, which may require treatment even after surgery, with advanced skeletal maturation, secondary or tertiary hypothyroidism and hypothalamic obesity syndrome [28]. Gelastic and dacrystic seizures are designated "mimetic automatisms" in the 2025 ILAE seizure classification [28].

In the individual patient data meta-analysis of 517 patients by Niazi and colleagues, Delalande types II or III were present in 215 of 316 (68.0%), gelastic seizures in 456 of 496 (91.9%), other focal or generalised seizures in 302 of 455 (66.4%), cognitive impairment in 188 of 297 (63.3%), neuropsychiatric comorbidity in 133 of 254 (52.4%) and preoperative precocious puberty in 92 of 313 (29.4%) [107]. In the systematic review by Luisi and colleagues (46 articles, 1318 patients), there was delay or intellectual disability in 424 of 819 (51.7%) and psychiatric disorders in 257 of 465 (55.3%) [108]. In the international survey by the Hope for HH association (256 respondents, mean age 18 years, self-reported data), more than 40% reported precocious puberty, 40% altered thermoregulation, 20% hypothyroidism and more than 40% a diagnosis of anxiety [17]. In the Freiburg cohort (n=41), patients had less slow-wave sleep than controls [17].

3.7. Natural history and treated disease

The natural history of untreated PHS includes neonatal death from adrenal insufficiency, which is preventable with recognition and cortisol replacement; Smith's states that death before 3 years of age "is not uncommon", whereas GeneReviews considers that, in the absence of potentially fatal malformations, the prognosis of non-familial PHS should be presumed to be good [1, 42]. The surgical outcome data for hamartoma (sections 7 and 11) refer to treated patients, largely non-syndromic. In GCPS, the sources describe large families with a mild form, good general health and normal longevity, and the data on function after limb surgery are descriptive [2, 7].

4. Clinical Recognition and Warning Signs

4.1. Prenatal period

Ultrasound detects a well-formed extra digit from the second trimester onwards, and other sonographic markers help confirm a syndrome [13]. In GCPS, polydactyly, macrocephaly or central nervous system anomalies (hydrocephalus) may be detected, but sonographic macrocephaly is poorly specific and a normal ultrasound does not exclude the diagnosis [2, 7]. In PHS, the reliability of ultrasound is unknown and a normal ultrasound in an at-risk fetus also does not exclude the diagnosis [1]. High-resolution fetal MRI allows the hamartoma to be detected, but the lumen of the third ventricle is only established at 26–28 weeks, which hampers diagnosis before the end of the second trimester; interhypothalamic adhesions are a cause of false diagnosis [17]. Prenatal diagnoses of GCPS in Spain and Portugal and of PHS in Brazil, the latter with ultrasound, MRI and three-dimensional reconstruction, have been published [109–111], as has a case of a GLI3 variant with short, hypomineralised bones, renal agenesis and heart disease [112]. In a Chinese cohort of 44 fetuses with polydactyly, the yield of genomic testing was 9.5% in isolated polydactyly and 60.9% in polydactyly associated with other anomalies, with one diagnosis in GLI3 (section 3.5) [104]. In the French series, the prenatal selection criteria for PHS were hamartoma or imperforate anus with at least two of five features (intrauterine growth restriction, limb malformation, heart disease, micropenis, renal anomaly) [51].

4.2. Newborn: emergency signs

In a newborn with mesoaxial or postaxial polydactyly associated with imperforate anus, micropenis, hypospadias, hydrometrocolpos, respiratory distress or aspiration, PHS should be considered immediately [1, 113]. The emergency is cortisol deficiency: adrenal crisis can be lethal in infants who are not assessed or treated, and assessment is urgent when there is no family history or when there are relatives with PHS and cortisol deficiency [1]. Neonatal hypoglycaemia, haemodynamic instability and the need for glucose infusions are warning signs of hypopituitarism [42]. Signs of aspiration or stridor require urgent otorhinolaryngological assessment to exclude a laryngotracheal cleft [1]. In the presence of anal anomalies, examination of the limbs guides the diagnosis (polydactyly in PHS, triphalangeal thumb in Townes-Brocks syndrome, radial anomalies in the VACTERL association) and justifies endocrine referral; genitourinary anomalies account for 49% of the anomalies associated with anal anomalies and cardiovascular anomalies for 27%, which justifies renal ultrasound and echocardiography [113]. There is no newborn screening for GLI3-related disorders.

4.3. Infant and child

In GCPS, suspicion arises in the presence of preaxial polydactyly of the feet or broad thumbs and halluces with cutaneous syndactyly, especially if there is macrocephaly, a high forehead and widely spaced eyes, or in the presence of a parent with operated polydactyly; polysyndactyly is a clue to GCPS in the assessment of hypertelorism [2, 68, 114]. Trigonocephaly due to metopic synostosis with polysyndactyly should raise the possibility of GCPS due to GLI3 deletion, and not only Carpenter syndrome [101]. Accelerated growth of head circumference, neurological signs, developmental delay or seizures are indications for MRI and, if there is significant delay, chromosomal microarray to exclude contiguous gene deletion [2].

In hypothalamic hamartoma, gelastic seizures have been present for longer than parents report, and video recorded on a mobile phone and video-electroencephalography (EEG) help to recognise them [28]. The 2026 international consensus considers a hamartoma likely when seizures are gelastic or dacrystic and when there are endocrine symptoms; gelastic seizures are not pathognomonic, because they may have a temporal or frontal origin [28]. Central precocious puberty is defined by the appearance of secondary sexual characteristics before 8 years of age in girls and 9 years in boys [29]. The 2026 Endocrine Society guideline exempts from watchful waiting, and recommends immediate assessment by paediatric endocrinology for, girls with premature thelarche who have central nervous system signs (headaches, seizures or visual field deficit) [29]. In a child with PHS or a known hamartoma, precocious puberty is by definition of organic cause.

4.4. Adolescent and adult

Mild forms of PHS may be identified in adulthood, in connection with polydactyly operated on in childhood, seizures, an imaging finding or family investigation [81, 82]. A family history of polydactyly in an adult with hypothalamic hamartoma or bifid epiglottis should prompt GLI3 testing [1, 88]. A bifid epiglottis identified at laryngoscopy should prompt a search for other PHS anomalies, which are often asymptomatic [88].

Table 4.1 — Warning signs by age

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

Table 5.1 — Diagnostic criteria

The GCPS GeneReviews states that there are no published consensus clinical criteria [2]. Al-Qattan and colleagues recommended revising the PHS criteria [54]. The findings suggestive of PHS are hypothalamic hamartoma, mesoaxial polydactyly, postaxial polydactyly (types A and B), bifid epiglottis, imperforate anus, renal anomalies (cystic malformations, hypoplasia, ectopic ureteral insertion), genitourinary anomalies (hydrometrocolpos), pulmonary segmentation anomalies, non-polydactylous skeletal anomalies and a compatible family history, the absence of which does not exclude the diagnosis [1]. The findings suggestive of GCPS are macrocephaly, widely spaced eyes with an interpupillary distance above the 97th centile, preaxial polydactyly with or without postaxial polydactyly, cutaneous syndactyly and an autosomal dominant family history [2]. Postaxial polydactyly type B, with a prevalence of about 1% in West and Central Africa, should be assessed with caution as a criterion in these populations [1, 2].

The 2010 NIH series applied inclusion criteria by group, assessed in sequence (first the oral-facial-digital overlap, then PHS and GCPS and finally "sub-PHS" and "sub-GCPS"; those meeting both of the latter were included in the group with more features) [24].

Table 5.2 — Inclusion criteria of the 2010 NIH series

Source: [24].

Johnston and colleagues distinguish the use of the criteria in the two syndromes: when the PHS criteria are met, the clinical diagnosis is quick and secure and allows potentially life-saving care to be instituted, molecular analysis being useful but not necessary to initiate it; in GCPS, the broadened criteria serve to select patients for molecular testing, and not for a clinical diagnosis without molecular data, and, as GCPS has no potentially fatal complications, molecular confirmation can be awaited [25]. The specificity of the findings is uneven: in 2005, preaxial polydactyly had been described in at least 25 syndromes and hypertelorism in at least 68, whereas hypothalamic hamartoma had been described in only 4 and insertional polydactyly in 6 [25].

5.2. Yield of molecular testing

Table 5.3 — Proportion of probands with a detected variant, by method

These proportions measure the percentage of probands with a variant among clinically selected patients at referral centres; they are not clinical sensitivity and do not apply to patients with partial phenotypes. The 2010 NIH series recommends extending the indication for testing to patients with one PHS or GCPS feature and one oral-facial-digital feature [24].

5.3. Tests in order

In PHS, the options are single-gene testing (GLI3 sequencing for missense, nonsense, splicing variants and small insertions and deletions), a multigene panel or, when the diagnosis has not been considered, exome or genome sequencing; large exonic deletions and duplications have not been described in PHS [1]. In GCPS, option 1 begins with GLI3 sequencing and, if negative, proceeds to targeted deletion/duplication analysis, with karyotyping if single-gene testing is negative; option 2 is exome or genome sequencing when the phenotype is indistinguishable from other polydactylies or macrocephalies [2]. GeneReviews suggests a multigene panel as a way of limiting the identification of variants of uncertain significance and of variants in genes that do not explain the phenotype, noting that the genes included and the sensitivity vary between laboratories; targeted deletion/duplication analysis may use exome or genome data in some laboratories [2]. When there is significant developmental delay or intellectual disability, or recurrent miscarriages in the parents, chromosomal microarray should be the first test [2]. The 2005 algorithm (G-banded karyotype, GLI3 FISH, sequencing and quantitative PCR, at a cost of at least 3000 dollars at the time) is historical and has been replaced by next-generation sequencing with copy-number analysis [2, 25]. High-density targeted microarray delimits the extent of deletions better than FISH [65]. The Polish series of Jamsheer and colleagues recommends including multiplex ligation-dependent probe amplification (MLPA) or quantitative polymerase chain reaction (PCR) in routine practice [100].

Brain MRI is the imaging test for the hamartoma: a non-enhancing mass in the floor of the third ventricle, posterior to the optic chiasm, isointense to grey matter on T1 and T2, which may have a distinct intensity on the FLAIR (fluid-attenuated inversion recovery) sequence; cranial CT and ultrasound are not adequate for diagnosis [1]. The 2026 consensus considers important a 3 T MRI with the HARNESS epilepsy protocol (3D T1 with a voxel of 1 mm³ or less, 3D FLAIR and T2 in two planes), 1.5 T being acceptable but 3 T preferred; this was the only item with unanimous agreement [28]. If gelastic seizures are clear and the MRI is negative, the MRI should be reviewed and repeated at 3 T [28]. Surface EEG may be normal: in a series of 133 patients, 75% of 584 recorded gelastic seizures (in 56% of patients) had no ictal EEG change, and the ictal EEG may falsely suggest a cortical focus [17]. Routine EEG, magnetoencephalography, intracranial EEG, 3D source localisation, CT, fluorodeoxyglucose positron emission tomography and ictal single-photon emission computed tomography were not considered important in the initial diagnosis by the 2026 consensus, which considered video-EEG important, with the caveat of false negatives and false localisation or lateralisation [28]. The consensus adds that paediatric guidelines require T2 in two planes, so T2 is mandatory in any child with suspected hamartoma (the lesion is best seen on this sequence), that the neuroradiologist must look specifically for the hamartoma, that new suggestive clinical data should prompt review of previous imaging and neurophysiology, and that stereoelectroencephalography (SEEG), although it confirmed that the hamartoma generates gelastic seizures, has no role in diagnosis or management [28]. Gelastic seizures have no surface EEG correlate in about 75% of patients, and those that do often localise falsely, so EEG should not influence the surgical decision and video may be diagnostic [28]. The 2014 ILAE Florence consensus, cited by the 2021 review, also advises against positron emission tomography and single-photon emission tomography in hamartoma [17, 116].

The Delalande classification describes the insertion of the hamartoma: type I, horizontal insertion below the floor of the third ventricle; type II, vertical insertion in the wall of the third ventricle, above the floor; type III, horizontal and vertical insertions, above and below the floor; type IV, giant hamartoma (by the Barrow criterion, a volume of 8 cm³ or more by the ellipsoid formula) [17].

The complementary assessment includes limb radiographs (to distinguish postaxial from central polydactyly and identify the Y-shaped metacarpal), laryngoscopy, renal ultrasound, assessment of the hypothalamic-pituitary axis and, in GCPS, measurement of head circumference [1, 2]. For the corticotroph axis, the 2016 Endocrine Society guideline on adult hypopituitarism suggests serum cortisol at 8–9 am as the first test (2|⊕OOO) and advises against random cortisol (1|⊕⊕OO); a value below 3 µg/dL indicates adrenal insufficiency and a value above 15 µg/dL makes it unlikely; between 3 and 15 µg/dL it suggests a corticotrophin stimulation test, in which a peak cortisol below 18.1 µg/dL (500 nmol/L) at 30 or 60 minutes indicates insufficiency; and samples should be taken at least 18–24 hours after the last dose of hydrocortisone [117]. These thresholds were defined in adults and are assay-dependent. For the thyrotroph axis, central hypothyroidism is diagnosed with combined FT4 and TSH: FT4 below the lower limit with low, normal or slightly elevated TSH confirms it (on at least two measurements, according to the ETA), and dynamic TSH testing is not recommended [27, 117]. The 2018 European Thyroid Association (ETA) guideline recommends screening for central hypothyroidism in children with GH deficiency, delayed or precocious puberty, growth or developmental delay and hypothalamic-pituitary lesions, and in newborns with hypotonia, prolonged jaundice or micropenis with cryptorchidism; in the follow-up of hypothalamic-pituitary disease, it suggests suspecting it when FT4 is in the lower quartile or falls by more than 20% on the same assay [27]. GH deficiency requires a stimulation test, because an isolated measurement is not useful, except when there is a clear clinical picture and three other documented pituitary deficiencies; central hypothyroidism should be treated before the test [117]. When central diabetes insipidus (arginine vasopressin deficiency) is suspected, serum and urine osmolality are measured simultaneously when urine output exceeds 50 mL/kg/24 hours [117]. For adrenal stimulation testing, the 2016 Endocrine Society guideline on primary adrenal insufficiency, written for primary adrenal insufficiency and applicable to PHS only by extension, suggests the standard intravenous corticotrophin dose of 250 μg in adults and children aged 2 years or more, 15 μg/kg in infants and 125 μg in children under 2 years, but recommends treating with hydrocortisone before the result if there are severe symptoms or crisis [115]. In precocious puberty, the 2026 guideline suggests starting the assessment with basal luteinising hormone (LH) on an ultrasensitive assay (capable of detecting concentrations below 0.05–0.1 IU/L), ideally sampled between 8 and 10 am, reserving stimulation testing with GnRH or an agonist, or repetition of morning LH, for when basal LH is low with suggestive clinical features [29]. Radiographs of the hands and feet are essential to identify the skeletal elements and minor signs, including in broad thumbs, which can only be assessed reliably on radiography in most affected individuals [68].

5.4. Limitations of the tests

A negative result does not exclude the diagnosis. In PHS, about 5% of patients with typical clinical features have no variant identified, which is consistent with another locus, such as SMO in recessive forms, or with cryptic GLI3 variants (deep intronic or mosaic) [1, 86]. Stoll and colleagues described in 2001 a patient with a PHS phenotype without a variant detected by the methods of the time [98]. Targeted deletion/duplication analysis detects from one exon to the whole gene, but may not detect the breakpoints of large deletions or the deletion of neighbouring genes; chromosomal microarray is needed to delimit the deletion, and this sizing has prognostic value [2, 7]. Karyotyping is needed for balanced translocations and inversions, which sequencing and microarray may not detect [2, 60]. Mosaicism in leucocytes may escape conventional sequencing, and low-level mosaicism in a parent may go unnoticed, as in family G068 of the French series [1, 51]. In non-syndromic hamartoma, somatic variants are not detectable in blood and molecular diagnosis requires the resected tissue, with allele fractions between 7% and 54% [59, 75, 76]. GLI3 RNA has low expression in whole blood, so the study of splicing variants may require fibroblasts [21].

5.5. Variant classification and reporting

The report should state the gene, the transcript and version (NM_000168.6), the HGVS (Human Genome Variation Society) coding and protein notation, the zygosity, the class according to the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria and the evidence used. For diagnosis, the "pathogenic" and "likely pathogenic" classes are treated in the same way [1, 2]. Novel frameshift and nonsense variants in the PHS region pose a particular problem: they are frequently novel and clinical and molecular data to classify them are lacking, and, as the PHS mechanism does not involve NMD, applying the ClinGen specification of the PVS1 criterion may classify them as variants of uncertain significance (VUS); GeneReviews recommends expert consultation, and the author of the chapter makes himself available for consultation on atypical variants [1]. A VUS neither establishes nor excludes the diagnosis [1, 2]. Familial segregation and the phenotype (hamartoma, mesoaxial polydactyly, bifid epiglottis) help with reclassification. Notations from articles prior to 2005 should be converted in VariantValidator, as shown by the variant of Ng and colleagues, described as c.3385_3386delTT and normalised as c.3386_3387del [57, 79]. Protein notation errors also occur in recent publications, as with the variant c.2151del, published with the protein "p.(Gln71HisfsTer16)" and normalised as p.(Gln717HisfsTer16) [57, 82].

5.6. Algorithm

Table 5.4 — Practical algorithm

5.7. Prenatal and preimplantation diagnosis

When the familial variant is known, prenatal testing (amniocentesis or chorionic villus sampling) and preimplantation genetic testing (PGT) are possible; in PHS, prenatal testing allows immediate postnatal treatment of adrenal insufficiency to be prepared [1, 2, 7]. In GCPS due to a balanced parental 7p14.1 rearrangement, both tests are equally possible [2]. Variants that are difficult to interpret require caution in prenatal diagnosis [7]. Both GeneReviews chapters note that there may be differences in perspective among professionals and within families regarding prenatal testing (and, in GCPS, preimplantation testing), which most centres consider a personal decision [1, 2]. Maeda-Usui and colleagues discussed the potential benefit of rapid genetic testing in PHS [118].

6. Differential Diagnosis

6.1. Approach

Polydactyly is associated with more than 100 disorders and more than 100 genes, among which GLI3 and SHH are the most frequently cited [7, 21]. Congenital differences of the upper limb of all causes have an incidence of 20–30 per 10,000 live births [119]. The approach begins with a three-generation pedigree with a search for minimal signs in relatives, consanguinity (which points to ciliopathies, DOOR syndrome — deafness, onychodystrophy, osteodystrophy and intellectual disability — Carpenter syndrome or acrocallosal syndrome), maternal age (trisomy 13), maternal diabetes and exposures in early pregnancy [13]. The examination describes and counts the digits (an outline of the hand or foot drawn on paper helps), distinguishes a complete postaxial digit from a nodule, assesses the thumb and hallux (number of phalanges, duplication, width, opposition) and looks for mesoaxial polydactyly, short stature, overgrowth, obesity, skull shape and size, ocular anomalies, oral frenula, lingual hamartomas, clefts, a narrow chest, heart disease and renal and genital anomalies [13]. Isolated unilateral polydactyly may be sporadic, whereas bilateral changes of the hands and feet are probably genetic and usually dominant [13]. Postaxial polydactyly combined with preaxial or mesoaxial polydactyly and/or syndactyly points to a syndrome [13]. The Oxford Desk Reference defines postaxial polydactyly, by mistake, as "radial or fibular"; the postaxial side is the ulnar side in the hand and the fibular side in the foot [1, 13].

The differential of polydactylies and brachydactylies due to other genes (linked in Orphanet to group ORPHA:2913, non-syndromic polydactyly) is broad; only the entities that are confused with GLI3 disorders are presented here.

6.2. Classification of preaxial polydactylies and syndactylies

Table 6.1 — Temtamy and McKusick classification of preaxial polydactyly

Sources: [13, 21].

Table 6.2 — Classification of syndactylies (Temtamy and McKusick), with relevance to the differential

Source: [58].

6.3. Differential diagnosis table

Table 6.3 — Differential diagnosis of GLI3-related disorders

6.4. Differentials by finding

Table 6.4 — Bifid epiglottis, laryngeal cleft, mesoaxial polydactyly and tibial hemimelia outside GLI3 disorders

Table 6.5 — Broad thumbs and halluces: differential of GCPS

If it is difficult to decide whether the thumb is broad, the sign is of no use as a diagnostic clue; the degree of involvement varies, especially in autosomal dominant syndromes, and a bifid thumb may be described as broad, so radiography is necessary in most affected individuals [68].

Table 6.6 — Differential of the VACTERL association in the Brazilian book of genetic syndromes

Source: [122]. About 90% of VACTERL cases appear sporadic, with a negligible recurrence risk; Hall and colleagues (1989) proposed, for a secure diagnosis, at least one anomaly in each of the three regions (limbs, thorax, pelvis and lower abdomen) and, for a probable diagnosis, at least two anomalies in each of two of these regions, in an association with an incidence of 1.6/10,000 [113, 122].

6.5. Main differentials

The most frequent differential of PHS is Bardet-Biedl syndrome, which is much more common than PHS; an atypical Bardet-Biedl syndrome resembling PHS may be more frequent than GLI3 PHS itself, and in cases of overlap the BBS genes should be tested [1]. The distinction rests on the retinal dystrophy, obesity and renal disease of Bardet-Biedl, versus the hamartoma, mesoaxial polydactyly and bifid epiglottis of PHS. The phenocopy described by Wang and colleagues, with Leber amaurosis due to CEP290 and polydactyly due to GLI3, shows that two molecular diagnoses can mimic a syndromic ciliopathy [105].

The recessive form due to biallelic SMO variants, which OMIM designates Pallister-Hall-like syndrome (241800), is the closest genetic differential of PHS. Rubino and colleagues described two siblings, aged 6 and 3 years, with hamartoma and polydactyly or syndactyly, compound heterozygous for a 2 bp deletion and a deletion including part of SMO: the elder had microcephaly, short stature, a pituitary cyst and Chiari type I malformation, with gelastic seizures controlled with medication and normal development; the younger had dacrystic seizures and language delay, and the hamartoma compressed the chiasm and was partially resected [20, 87]. Le and colleagues extended the description to seven patients from five families, all with postaxial polydactyly, three with hamartoma and gelastic or dacrystic epilepsy, three with atrioventricular septal defect, one with short-segment Hirschsprung disease, and several with craniofacial dysmorphism, a narrow chest and short long bones; Green and colleagues described it as a recessive cause of PHS [20, 86]. The older descriptions of congenital hypothalamic hamartoma recurring in siblings (Marcuse 1953; Encha-Razavi 1992, who proposed the designation "congenital hypothalamic hamartoma syndrome") are also grouped under this entry [20]. The distinction has consequences for counselling: a 25% risk for the siblings of a case with healthy parents, versus a low risk, conditioned by germline mosaicism, in a de novo GLI3 case [1, 86].

In GCPS, the main differentials are acrocallosal syndrome and Carpenter syndrome. The mild end of KIF7 acrocallosal syndrome overlaps with the severe end of GCPS, and the distinction between GCPS-CGS (de novo or dominant) and acrocallosal syndrome (recessive) changes the recurrence risk for the parents from less than 1% to 25% [2, 7]. In Carpenter syndrome, craniosynostosis is constant and inheritance recessive; three children with complete GLI3 deletion and trigonocephaly had been labelled as Carpenter syndrome [101]. Carpenter corresponds to acrocephalopolysyndactyly type II; type I (Noack) is now considered Pfeiffer syndrome, and the Summitt and Goodman forms (type IV) lie within the clinical spectrum of Carpenter [55]. In the French series, the macrosomia of at least 13% of patients with GCPS raises the differential with Simpson-Golabi-Behmel syndrome type 1 (GPC3) [51]. Craniofrontonasal syndrome has, in women, a face similar to that of GCPS, and X-linked inheritance with females more affected is distinctive [2].

In imperforate anus with digital anomalies, Townes-Brocks syndrome (triphalangeal thumb, dysplastic ears and deafness) and the VACTERL association (radial, vertebral and cardiac anomalies) are distinguished from PHS by the absence of hamartoma and bifid epiglottis and by the type of polydactyly [113, 122]. Holoprosencephaly with polydactyly suggests PHS or pseudotrisomy 13 (holoprosencephaly-polydactyly syndrome, autosomal recessive, with postaxial polydactyly and heart disease, similar to hydrolethalus syndrome); in pregnancy, trisomy 13 is excluded by QF-PCR (quantitative fluorescent polymerase chain reaction), and examination of the parents should include testing for anosmia, because autosomal dominant holoprosencephaly has incomplete penetrance and extreme variability, with transmitting parents who may have only a single central incisor or be clinically normal [93]. In congenital hypopituitarism with a midline anomaly, septo-optic dysplasia is the main endocrine differential, and MRI distinguishes optic nerve and pituitary hypoplasia from hamartoma [103]. The overlap between PHS and oral-facial-digital syndrome type VI (Y-shaped metacarpal) led to the proposal that GLI3 should be studied in oral-facial-digital syndrome only when there is a pathognomonic sign of PHS (hamartoma, mesoaxial polydactyly, bifid epiglottis or imperforate anus) [51].

In agenesis of the corpus callosum, present in some patients with GCPS, the Oxford Desk Reference includes in the differential the tubulinopathies (TUBA1A, TUBB2A, TUBB2B, TUBB3, TUBB, TUBG1; biallelic TUBA8), Coffin-Siris-type conditions (SWI/SNF complex: ARID1B, SMARCA4, SMARCB1), acrocallosal syndrome, CEDNIK syndrome (SNAP29), GCPS due to large deletion (overlapping with acrocallosal syndrome), mosaic trisomy 8, Mowat-Wilson syndrome (ZEB2), FOXG1 and ARX; investigation includes microarray, panel, exome or genome, ophthalmological assessment, urinary amino acids and organic acids, cerebrospinal fluid glycine and lactate if there are seizures, and very-long-chain fatty acids in a boy [103]. Corpus callosum fibres begin to cross the midline at 11 weeks and the structure acquires its adult shape at 18–20 weeks, which limits early sonographic diagnosis [103].

In the syndactylies, the examination records the digits involved, whether the fusion is complete or partial, bony or soft-tissue only, shortening (symbrachydactyly of the Poland anomaly), symmetry, the thumbs and halluces, the skull (Apert syndrome), microphthalmia with a "pinched" nose (oculodentodigital dysplasia), clefts, frenula and lingual cysts (oral-facial-digital syndromes) and streaky skin pigmentation (chromosomal mosaicism), and examines the parents [58]. Asymmetric syndactylies suggest the Moebius sequence (congenital palsies of cranial nerves VI and VII with terminal transverse defects, sometimes with the Poland anomaly; the recessive Carey-Fineman-Ziter phenocopy is accompanied by clubfoot and muscle weakness) or amniotic bands (terminal syndactyly, in which a probe passes between the fingers at the base; recurrence below 2%) [58].

7. Treatments and Risk Management Strategies

7.1. Principles and sources of recommendations

There is no therapy directed at GLI3 or at the hedgehog pathway for GLI3-related developmental disorders; treatment is symptomatic and organ-based [1, 2, 7]. Nor are there clinical practice guidelines specific to GCPS or PHS: the recommendations of the GCPS GeneReviews are based on the personal experience of the authors, and GeneGuidance classifies the treatment evidence as expert reviews, without a ClinGen actionability assertion [2, 23]. The GCPS GeneReviews recommends supportive care, ideally multidisciplinary, to improve quality of life, maximise function and reduce complications [2]. For problems shared with other disorders, the general guidelines apply: the 2026 international Delphi consensus on hypothalamic hamartoma [28], the 2026 Endocrine Society guideline on central precocious puberty [29], the 2016 Endocrine Society guideline on primary adrenal insufficiency [115], the 2016 Endocrine Society guideline on hormone replacement in adult hypopituitarism [117], the 2018 ETA guideline on central hypothyroidism [27] and the 2016 Pediatric Endocrine Society guideline on GH in children, of which only the abstract and a review by the authors were consulted [131, 132]. None of them is specific to PHS, and their application is an explicit extrapolation. GeneGuidance identifies two interventions with an impact on PHS: recognition of neonatal cortisol deficiency and refraining from resection of the hamartoma [23].

Table 7.1 — Initial assessment after diagnosis (GeneReviews)

7.2. Adrenal insufficiency and hypopituitarism in PHS

Cortisol deficiency is the most urgent endocrine manifestation and the first to be assessed and treated; the remaining pituitary deficiencies are treated as in the general population, after assessment and treatment of corticotrophin deficiency [1]. The original 1980 series shows what happens without treatment: the six infants had hypoadrenalism and died in the neonatal period [41]. The adrenal insufficiency of PHS is central (secondary to hypopituitarism), with preserved aldosterone. The 2016 guideline on adult hypopituitarism recommends against fludrocortisone in secondary adrenal insufficiency (1|⊕⊕⊕O) and considers stress doses similar to those of the primary form; the paediatric doses come from the primary adrenal insufficiency guideline and are applied here by extension, without sodium chloride supplementation or fludrocortisone [115, 117]. In children, the guideline for the primary form suggests hydrocortisone in three or four doses, with a total initial daily dose of 8 mg/m² (2|⊕⊕○○), avoiding long-acting synthetic glucocorticoids (prednisolone, dexamethasone) and monitoring clinically (growth velocity, weight, blood pressure, energy) [115].

Table 7.2 — Hydrocortisone: replacement and stress doses in children (central adrenal insufficiency, by extension of the guideline for the primary form)

The adverse effects of glucocorticoid overdose include growth delay, Cushing syndrome, reduced bone mineral density, immunosuppression with risk of infection and psychiatric reactions; the US label recommends the lowest effective dose, monitoring of growth and a search for signs of Cushing syndrome every 6 months, with more frequent monitoring before 1 year of age [133]. Underdosing, abrupt discontinuation and switching between oral hydrocortisone formulations can precipitate adrenal crisis and death, and stressful situations (infections, surgery) require dose increases [133]. The 2016 guideline recommends educating the patient and family about stress doses and providing a steroid card and a glucocorticoid preparation for emergency parenteral administration [115]. If hydrocortisone is unavailable, prednisolone is the suggested alternative and dexamethasone the least preferred [115].

Table 7.3 — Hydrocortisone in the adolescent who has completed growth and in the adult with central adrenal insufficiency

The 2016 guideline suggests educating patients who take hydrocortisone and start enzyme-inducing antiseizure medications about the early signs of adrenal insufficiency (2|⊕⊕OO): phenytoin, carbamazepine, oxcarbazepine and topiramate accelerate the hepatic metabolism of the hormones, the effect on hydrocortisone is smaller than on dexamethasone, but oxcarbazepine increases cortisol clearance, and the dexamethasone dose should be increased if inducers are combined (2|⊕OOO) [117]. This interaction is relevant in PHS and hamartoma, because oxcarbazepine or carbamazepine were the most voted first-line antiseizure medications in the 2026 consensus (section 7.5) [28].

7.2.1. Levothyroxine in central hypothyroidism

The 2018 ETA guideline recommends starting levothyroxine only after cortisol secretion has been confirmed, and, if central adrenal insufficiency has not been excluded, starting the glucocorticoid first, because of the risk of precipitating a crisis (1|⊕⊕⊕⊕); the 2016 adult guideline suggests the same and, if assessment is not possible, empirical glucocorticoid until definitive assessment (2|⊕⊕OO) [27, 117]. Titration is guided by FT4, with a target above the median of the reference range (ETA) or in the upper half (Endocrine Society); TSH is not useful for adjusting the dose when it was low at baseline (ETA, 1|⊕⊕OO; Endocrine Society, 1|⊕⊕⊕O), and the adult guideline advises against liothyronine and thyroid extracts (2|⊕⊕OO) [27, 117].

Table 7.4 — Levothyroxine in central hypothyroidism: doses and monitoring

The US label titrates by TSH and/or free T4 in all forms, whereas the ETA titrates central hypothyroidism by FT4 only; the paediatric doses of the two sources overlap [27, 137]. Levothyroxine is taken fasting, 30 to 60 minutes before breakfast, and in children the tablet may be crushed in 5–10 mL of water; biotin should be stopped at least 2 days before blood tests [137]. The dose should be increased when GH is started, with oestrogens or oral contraceptives, with pubertal development, in pregnancy, with weight gain and with drugs that alter absorption or metabolism, including antiseizure medications; the adult guideline recommends measuring FT4 at least 6 weeks after starting an antiseizure medication (1|⊕⊕OO) and preferring an equilibrium dialysis FT4 assay with phenytoin [27, 117]. Undertreatment is suggested by FT4 below or near the lower limit, especially with TSH above 1.0 mU/L and compatible clinical features [27].

7.2.2. Somatropin in growth hormone deficiency

The 2016 Pediatric Endocrine Society guideline, the first on GH developed using the GRADE method (Grading of Recommendations, Assessment, Development and Evaluation), recommends with high-quality evidence GH to normalise adult height and avoid extreme short stature in children and adolescents with GH deficiency, advises against the use of GH stimulation tests as the sole diagnostic criterion and recommends monitoring for intracranial hypertension, slipped capital femoral epiphysis and progression of scoliosis; it also recommends reassessing the adrenal and thyroid axes after starting GH in patients with possible multiple pituitary deficiencies and states that the data are insufficient to know whether GH increases the risk of neoplasia in patients with genetic syndromes of increased baseline risk [131, 132]. The full text of the guideline was not consulted, so the doses below come from the labels.

Table 7.5 — Somatropin and long-acting analogues in paediatric GH deficiency

Minimum ages differ between jurisdictions (Sogroya®: 2.5 years in the USA and 3 years in the EU; Skytrofa®: 1 year and 11.5 kg in the USA and 3 years in the EU), and the additional indications of Sogroya® (small for gestational age, Noonan syndrome, idiopathic short stature) appear only in the US label [141, 142, 145, 146]. In adults with proven GH deficiency, the 2016 guideline recommends starting with 0.2–0.4 mg/day before 60 years of age and 0.1–0.2 mg/day thereafter (1|⊕⊕⊕O), increasing by 0.1–0.2 mg/day at intervals of about 6 weeks, keeping IGF-I below the upper limit of normal (1|⊕OOO), giving higher doses to women on oral oestrogen (2|⊕⊕⊕O) and stopping GH in pregnancy (2|⊕⊕OO) [117]. GH can unmask central hypothyroidism and adrenal insufficiency: the adult guideline suggests testing the corticotroph axis before and after starting GH (2|⊕OOO) and monitoring FT4 (1|⊕⊕OO), and the ETA recommends reassessing thyroid function and increasing levothyroxine when GH is started [27, 117]. With GnRH agonists, GH is added only when there is GH deficiency with its own indication (recommendation 9 of the 2026 guideline) [29].

7.2.3. Desmopressin in arginine vasopressin deficiency (central diabetes insipidus)

Table 7.6 — Desmopressin: formulations and doses

Bioavailability is below 1% by the oral and sublingual routes and about 6% by the nasal route; the 2016 guideline gives as equivalent 60 µg of lyophilisate, 100 µg of tablet and 2.5 µg of nasal spray [117]. All labels warn of hyponatraemia and water intoxication, especially in children and the elderly, with fluid restriction, sodium monitoring and discontinuation in the presence of headache, nausea, vomiting, weight gain or seizures [147–150]. The 2016 guideline suggests individualised regimens, educating all patients about the risk of overdose and allowing at least once a week a phase of polyuria in which the drug effect wears off, carefully titrating desmopressin and fluid intake in adipsic diabetes insipidus, with frequent weighing and sodium measurement, and wearing an emergency bracelet or necklace (ungraded good practice statements) [117]. After pituitary surgery, it suggests short-acting aqueous antidiuretic hormone subcutaneously initially, avoiding scheduled doses in the first week because of the risk of hyponatraemia and of inappropriate antidiuretic hormone secretion at 7–10 days, oral or nasal desmopressin only if there is significant polyuria after discharge, and at least one attempt at withdrawal in the following weeks or months [117]. Applying these rules to the postoperative period after hamartoma surgery, in which hypernatraemia due to arginine vasopressin deficiency is described, is an extrapolation [28]. Carbamazepine, oxcarbazepine, lamotrigine, perampanel and felbamate can cause hyponatraemia by increasing the renal response to desmopressin, and the guideline suggests adjusting the dose when an antiseizure medication is started (2|⊕⊕OO) [117]. Adrenal insufficiency can mask partial diabetes insipidus, which may become evident after the glucocorticoid is started (2|⊕OOO) [117].

7.2.4. Sex steroids and other interactions

In adults with central hypogonadism, the 2016 guideline suggests testosterone in males (2|⊕⊕OO) and recommends oestrogen, combined with a progestogen if the uterus is present, in premenopausal females (1|⊕⊕⊕O), until the average age of natural menopause; it refers the testosterone formulations and doses to the specific Endocrine Society guideline, which was not consulted [117]. Oral oestrogens increase corticosteroid-binding globulin and total cortisol, require higher doses of GH and levothyroxine, and the oestrogen-progestogen combination can lower lamotrigine levels, so the guideline suggests reviewing antiseizure medication levels when oestrogen is started (2|⊕⊕OO) [117].

The sequence recommended by GeneReviews, with assessment of the corticotroph axis before the somatotroph, gonadotroph and thyrotroph axes, applies to all patients [1]. Fertility and pregnancy, uncommon in hypopituitarism due to PHS, follow the general guidelines: in a review of 31 pregnancies in 27 women with hypopituitarism, there was postpartum haemorrhage in 8.7%, transverse presentation in 16% and small-for-gestational-age newborns in 42.4% [1, 151]. In pregnancy, the 2016 guideline suggests hydrocortisone as the preferred glucocorticoid, with dose increases especially in the third trimester and clinical monitoring for under- and over-replacement, recommends against dexamethasone, because it is not inactivated by the placenta (1|⊕⊕OO), recommends a stress dose in active labour (50 mg intravenously in the second stage; 100 mg every 6–8 hours for caesarean section) (1|⊕⊕OO), FT4 or total T4 every 4–6 weeks with an increase in levothyroxine (1|⊕⊕OO), continuing desmopressin with adjustment (2|⊕⊕OO) and stopping GH (2|⊕⊕OO) [117].

7.3. Central precocious puberty

Central precocious puberty is the typical preoperative endocrinopathy of hypothalamic hamartoma and is frequent in the parahypothalamic hamartoma of syndromic forms [17]. The 2026 Endocrine Society guideline, developed using the GRADE method with systematic reviews by the Mayo Clinic (searches in February 2025, updated in January 2026), issued ten recommendations, all conditional and with very low or low certainty of evidence [29].

Table 7.7 — Recommendations of the 2026 Endocrine Society guideline and application to hypothalamic hamartoma

In the guideline's supporting systematic review, the prevalence of pathological MRI findings (tumours or hamartomas) was 11% (49/443) in girls under 6 years, 1% (22/2154) in girls aged 6.0 to 8.0 years, 28% (39/139) in boys under 8 years and 0% (0/337) in boys aged 8.0 to 9.0 years; the prevalence of hamartomas was 9% (41/443), 0.4% (8/2154), 13% (18/139) and 0% (0/337), respectively [29]. The review pooled 35 single-arm studies from 19 countries, with 5541 children with central precocious puberty (mean age 7 years; 83% girls), without comparative studies between performing and not performing MRI: there were MRI anomalies (pathological and incidental) in 18% and pathological lesions in 5.5%; the most frequent tumours were astrocytomas, craniopharyngiomas and optic gliomas, and it was not possible to know what proportion had neurological symptoms [29]. Hamartoma was the most frequent pathological lesion in girls under 6 years. These data measure the proportion of lesions in referred children with central precocious puberty, not the frequency of precocious puberty in hamartoma. Studies published in 2026 point in the same direction: in 1087 Chinese children with precocious or rapid puberty who underwent MRI, there were lesions in 14.5% of girls and 19.6% of boys, and the 26 pathogenic lesions (hamartomas or gliomas) occurred only in central precocious puberty, with earlier onset (1.9 years in girls and 1.0 year in boys) and higher LH, requiring higher agonist doses at 6 months in girls (p=0.025); the authors propose routine MRI in girls under 6 years and boys under 7 [152]. In Riyadh, among 70 girls with central precocious puberty, 43 underwent MRI and 17 had anomalies, 4 of them hamartomas [153]; in a Turkish series of 46 boys, 32% had a neurogenic cause and hamartoma accounted for 13% of the lesions [154].

In the evidence on treatment, which includes no randomised trials (15 observational studies from 12 countries, 1835 girls with onset at 8.2 ± 1.1 years), the agonist was associated with an adult height gain of 2.7 cm (95% CI 1.1–4.4), of 3.9 cm with 3 or more years of treatment and of 2.1 cm with less than 3 years, with no relevant effect on body mass index (0.4 kg/m²; −0.7 to 1.6), insufficient evidence on bone mineral density and menarche at 12.0–12.4 years versus 9.6–11.4 years in controls [29]. Starting after 7 years of age may not improve adult height, and with a bone age of 12 years in girls or 13 in boys no benefit is expected [29]. In the comparison of formulations (3 randomised trials and 7 observational studies, 759 participants), the proportion with LH suppression was equivalent (relative risk 1.00; 95% CI 0.94–1.07) [29]. There are no oral formulations, and the implant requires facilities for insertion and removal, sometimes under anaesthesia [29]. In a child with a known hamartoma, recommendations 1, 2 and 4 (watchful waiting and non-systematic MRI) do not apply, because the organic cause and imaging already exist; recommendations 3 and 6 to 10 apply, and recommendation 10 allows treatment to be prolonged in children with neurocognitive impairment, which is frequent in hamartoma with epilepsy (inference from the guideline) [29]. "Escape" from suppression, sometimes linked to inadequate dose or absorption, was estimated at about 8.9% of published cases of precocious puberty due to hamartoma; in a 15-month-old boy without complete suppression despite dose increase and change of formulation, the androgenic signs were controlled with an aromatase inhibitor and an antiandrogen, an off-label use without long-term safety data [155].

Table 7.8 — GnRH agonists for central precocious puberty (US labels)

The labels describe the same class adverse effects: initial rise in gonadotrophins and sex steroids, with transient worsening of pubertal signs and possible vaginal bleeding in the first weeks; psychiatric events (emotional lability, crying, irritability, impatience, anger, aggression); seizures, in patients with or without a history of epilepsy, central nervous system anomalies or tumours; severe cutaneous adverse reactions, which require interruption and, if confirmed, permanent discontinuation; and pseudotumour cerebri (idiopathic intracranial hypertension), with monitoring for headache, papilloedema and blurred vision [156, 158, 159]. Hypersensitivity to GnRH or to the agonists and pregnancy are contraindications [156–159]. The 2026 guideline adds sterile abscesses, which are rare and even rarer with recent formulations, and headaches [29]. In hamartoma with epilepsy, the seizure warning and psychiatric events (aggression, which is already frequent in hamartoma) require neurological and behavioural monitoring during treatment [17, 156].

In the EU, GnRH agonists for central precocious puberty do not have a centralised authorisation; they are authorised through national, decentralised or mutual recognition procedures, and the SmPCs differ between countries. The Portuguese SmPCs, from INFARMED (National Authority of Medicines and Health Products), were not consulted; table 7.9 uses the Spanish SmPCs from the public database of the Agencia Española de Medicamentos y Productos Sanitarios (AEMPS) as an example of European text [160–164].

Table 7.9 — GnRH agonists in central precocious puberty: example of national EU SmPCs (Spain)

The European SmPCs warn of seizures, "especially in women and children", including patients with epilepsy, intracranial tumours or central nervous system anomalies and also patients without risk factors, and, with leuprorelin, of idiopathic intracranial hypertension, with monitoring for severe or recurrent headaches, visual disturbances and tinnitus and consideration of discontinuation; bone mineral density may decrease during treatment, with peak bone mass apparently preserved [160, 162–164]. The 2026 guideline confirms the recent pseudotumour cerebri warning in the labels [29]. Starting ages and discontinuation criteria differ between products (before 8 or 9 years in girls), between the EU and the USA (Fensolvi® and Triptodur® from 2 years; Lupron Depot-PED® with weight-based doses for up to 25, 25 to 37.5 and more than 37.5 kg) and between the European SmPCs, which set discontinuation by bone age (12–13 and 13–14 years), and the 2026 guideline, which sets it by a chronological age of 10–11 and 11–12 years and/or a bone age of 11–12 and 12–13 years [29, 156–158, 160].

Table 7.10 — Comparative regulatory status

The 2026 guideline notes that the annual cost of GnRH agonists was about 1100 to 1300 dollars in France and Spain and 30,000 to 60,000 dollars (average wholesale price) in the USA in 2025 [29].

There is a divergence between the labels and the guideline: the US labels recommend hormonal monitoring of suppression 1–2 months after starting, and the 2026 guideline suggests not monitoring biochemically on a routine basis, reserving measurements for clinical suspicion of failure (progression of breast development or testicular volume, acceleration or persistence of growth velocity), after adherence and administration technique have been checked [29, 157, 158]. The previous consensus statements on GnRH agonists in children (that of the European Society for Paediatric Endocrinology and the Lawson Wilkins Pediatric Endocrine Society, of 2009, and that of the international consortium of 2019) have been partly superseded by the 2026 guideline [29, 165, 166].

7.4. Hypothalamic obesity

About 20% of patients with hamartoma develop hypothalamic obesity syndrome (hyperphagia and weight gain), especially after surgery; postoperative injury to the posterior hypothalamic subnuclei (dorsomedial nucleus and dorsal hypothalamic area) correlates with the syndrome, and premorbid obesity is present in up to half of patients before surgery [17]. The strategies described are diet and exercise, octreotide, metformin, possibly recent anti-obesity drugs and bariatric surgery; the sources consulted provide no dosage for this indication, which is off-label [17]. At Great Ormond Street Hospital, among 28 children treated with MRI-guided laser interstitial thermal therapy (2019–2024; median age 6 years), 15 of 26 were already overweight or obese before the intervention and 16 of 28 had increased appetite afterwards, less likely with ASD (p=0.049) and more likely when the ablation reached healthy hypothalamus (p=0.03); the median body mass index standard deviation score went from 1.485 to 1.89 at 12 months (n=15), and the authors recommend integrated dietetic support [167]. In a series of 169 patients treated with the same technique, there was transient hyperphagia in 26.7%, persistent obesity in 12% and permanent hypothyroidism in 1.8% [168].

7.5. Epilepsy associated with hypothalamic hamartoma

In PHS, seizures are treated symptomatically and generally respond to antiseizure medications; the hamartoma is removed or biopsied only in very unusual circumstances, because the complications and lifelong hormone replacement outweigh the benefits [1]. GeneReviews includes biopsy and resection of the hamartoma among the interventions to avoid [1]. Many patients with PHS and parahypothalamic hamartoma have only endocrinopathy and do not need surgery; the decision is individual [17]. In non-syndromic hamartoma, and in the rare patients with PHS and refractory epilepsy, the 2026 international consensus applies.

The 2026 modified Delphi consensus was conceived by the medical board of the Hope for HH association (paediatric neurosurgeons, paediatric epileptologists, neuropsychologists, endocrinologists and researchers) on the basis of focus groups with clinicians, researchers, patients and carers; the areas were defined at the 2019 international symposium (Washington), the first round was discussed at the 2022 symposium (Calgary) and the results were presented at the 2025 symposium (Barcelona) [28]. Of the 24 ILAE level II epilepsy surgery centres invited, 17 (70%) responded, from eight countries on five continents, with no African centres and with Asia and South America under-represented; the 32 respondents were neurosurgeons (12), paediatric neurologists (8), neurologists (4), neuropsychologists (4), neuroradiologists (3), one endocrinologist and one other professional, and the centres operated on a mean of 10.5 hamartomas per year (2 to 25) [28]. Consensus, defined as agreement of 75% or more on a 9-point Likert scale, was reached on 82% of the questions (round 1: 257 questions, no consensus on 45; round 2: 81 questions, no consensus on 2); the recommendations are expert opinion, without GRADE grading, and the authors acknowledge selection bias, a predominance of North American and European centres, a possible preference of each centre for its own techniques and non-response bias [28]. There are no formal guidelines for the diagnosis and treatment of hamartoma [28].

Most patients with non-syndromic hamartoma develop drug-resistant epilepsy beginning with gelastic seizures; the aim of antiseizure medications is to control seizures, prevent progression and limit generalised tonic-clonic seizures, and there are no data showing the superiority of any drug [28]. There was no consensus on the first-, second- or third-choice antiseizure medication (table 7.11); dosage follows that of focal epilepsy in general and is not specific to hamartoma [28]. The GCPS GeneReviews states that many antiseizure medications may be effective, but none has been shown to be specifically effective for seizures in GCPS, and recommends educating parents and carers about the common presentations of seizures [2]. In PHS with hypopituitarism, the choice of antiseizure medication should take into account interactions with hormone replacement: enzyme inducers (carbamazepine, oxcarbazepine, phenytoin, topiramate) accelerate the metabolism of levothyroxine and sex steroids, oxcarbazepine increases cortisol clearance, and carbamazepine, oxcarbazepine and lamotrigine potentiate desmopressin, with a risk of hyponatraemia (section 7.2) [117].

Table 7.11 — Votes of the 17 centres of the 2026 consensus by antiseizure medication in hypothalamic hamartoma (number of centres)

Source: [28]. No choice reached the consensus threshold; the empty cells of the article correspond to zero votes.

The consensus recommends starting the presurgical assessment with the first antiseizure medication and considering surgery after failure of the second appropriately chosen and dosed antiseizure medication; surgery should not be performed without a trial of antiseizure medication, waiting for three or more failures is too late and there was no consensus on operating at diagnosis or after failure of the first; early surgical assessment (before two antiseizure medications) is associated with better outcomes and possibly fewer comorbidities [28]. This position is aligned with the 2022 consensus of the ILAE Surgical Therapies Commission (61 experts from 28 countries), which recommends offering referral for surgical assessment to every patient with drug-resistant epilepsy as soon as drug resistance is established, regardless of duration, seizure type, location and comorbidities, and considering it, in adults and children, even with seizures controlled by one or two antiseizure medications when there is a lesion in non-eloquent cortex [169]. The 2026 consensus considers hamartoma surgery safe, with limited morbidity and no significant risk of mortality [28]. The surgical aim is to disconnect or completely remove the hamartoma, or to disconnect and partially remove it; disconnection can suppress seizures even with most of the hamartoma intact [28]. Laser interstitial thermal therapy (LITT) was considered appropriate for Delalande types 2 and 3, the skull base approach was considered inappropriate for type 2, and medical treatment alone was considered inappropriate for all four types; for Gamma Knife radiosurgery, radiofrequency thermocoagulation (RFTC), endoscopic disconnection and the transcallosal approach there was no consensus for any type, and there was no consensus on the minimum age for LITT, Gamma Knife and RFTC or on reoperation; for the endoscopic, transcallosal and skull base routes, the minimum age was considered to have no limit (table 7.12) [28].

Table 7.12 — Appropriateness of techniques by Delalande type in the 2026 consensus

Source: [28], table 4 re-extracted from the full text; the article does not explain the asterisk associated with "no limit". What matters is the surgeon's experience with the technique, the size and location of the hamartoma, the severity of the epilepsy and the expertise of the centre [28]. Of the 17 centres, 15 (88%) offered endoscopic or transcallosal resection, 14 (82%) skull base surgery, 12 (71%) LITT, Gamma Knife or RFTC and 3 of 15 (20%) high-intensity focused ultrasound; the endoscopic, transcallosal and skull base approaches and focused ultrasound did not require external referral, unlike LITT, Gamma Knife and RFTC [28]. After surgery, the consensus recommends a 3 T MRI with an epilepsy protocol between 6 and 12 months and further imaging only if seizures persist, endocrine referral of all patients at diagnosis, multidisciplinary endocrine follow-up coordinated with neurology, especially postoperatively (no consensus on which tests to perform), and baseline psychiatric assessment [28]. The 2021 review describes a trend towards minimally invasive techniques [17].

Table 7.13 — Outcomes of surgical treatment of epilepsy due to hypothalamic hamartoma (synthesis studies and series)

Synthesis studies compare techniques applied to different patients, by different centres and with different follow-up periods; the differences between techniques are indirect comparisons and do not prove superiority. With a seizure-freedom rate of about 50% after the first procedure and about 15% major complications, the choice depends on the relationship between the expected absolute benefit and the individual risk [107]. The predictors of lower success were multiple seizure types (OR 0.296; 95% CI 0.140–0.624) and previous surgery (OR 0.418; 0.198–0.884); compared with open microsurgery, the risk of major complications was lower with radiosurgery (OR 0.024), RFTC (0.133) and MRgLITT (0.234) [107]. The authors consider MRgLITT and RFTC first-line techniques and radiosurgery an option for small hamartomas with good baseline functioning [107]. Between 20% and 30% of patients need more than one treatment, and delay and psychiatric comorbidity may persist after successful removal [178].

The complications described by technique are, for endoscopic disconnection, up to 8.3% in 112 patients (transient memory deficit, transient or persistent third nerve palsy, hemiplegia, infection); for LITT, salt wasting in large hamartomas and, in a series of 18 patients, weight gain (22%), short-term memory loss (22%) and late hypothyroidism (11%); for the transcallosal approach (26 patients), transient memory dysfunction in 58% with behavioural improvement in 88% [17]. In the review by Luisi and colleagues, the most frequently reported complications were hyperphagia and weight gain, electrolyte disturbances, diabetes insipidus, hypothyroidism and transient hyperthermia or poikilothermia; memory deficits were more frequent after microsurgery, and hemiparesis and cranial nerve palsies after microsurgery or endoscopy [108].

The 2026 consensus summarises the individual data of Niazi and colleagues by technique (table 7.14): overall complications were 54.6% with open microsurgery and 1.8% with radiosurgery, and major complications 33% and 1.8%, respectively [28, 107].

Table 7.14 — Complications by technique in the individual data of Niazi 2024, according to the 2026 consensus

Source: [28]. The authors warn that the rates, including the many values of 0%, reflect study design and may underestimate reality through under-reporting, and that prospective and randomised data are lacking; the 2025–2026 series in table 7.13 describe, for example, transient hyperphagia in 26.7% after MRgLITT and persistent complications in 14.3% after RFTC [28, 168, 172]. In the Freiburg RFTC series, the coagulated volume did not correlate with seizure freedom, which suggests that disconnection matters more than volume reduction [172]. In 2026 an operative video of disconnection of a hamartoma by a trans-Sylvian approach in a patient with PHS was published, without outcome data [179].

7.6. Limbs

In GCPS, surgical correction of polydactyly is elective; preaxial polydactyly of the hands takes priority over postaxial polydactyly of the hand and over any polydactyly of the foot, because of the importance of early development of grasp [2]. In the feet, cosmetic and footwear benefits must be weighed against potential orthopaedic complications, and GeneReviews includes correction of the feet without careful consideration among the practices to avoid [2]. Cutaneous syndactyly that is more than minimal is generally corrected [2]. The 2008 review states that surgery for polysyndactyly often has a good outcome, that postaxial polydactyly is easier to repair than preaxial polydactyly (which should be operated on by experienced surgeons), that foot surgery is optional in many cases and that iatrogenic biomechanical consequences can be severe; one family with PAP-A chose not to operate on a well-formed, functional digit [7]. In the Lithuanian family in section 10, the hand operations took place between 1 and 4 years of age in the proband and, in the mother, from 6 months at 6-month intervals; the mother had iatrogenic pain after foot surgery [21]. In PHS, correction of mesoaxial polydactyly is more complex and should be performed by an experienced surgeon, with occupational therapy for dexterity in the presence of digital malalignment [1]. Occupational therapy (fine motor skills) and physiotherapy (gross motor skills), mobility, activities of daily living and adaptive devices complete the treatment [2]. The sources consulted do not include a consensus on the ideal age for surgery.

7.7. Airway, digestive tract, kidney and genitalia

Bifid epiglottis, which is often asymptomatic, requires no treatment, except in the case of obstruction or other anomalies (for example, tracheal stenosis); laryngeal clefts are treated according to type and respiratory compromise [1]. In one patient with cricoid stenosis, tracheostomy was required; the same patient underwent perineal anorectoplasty and vesicostomy [98]. Imperforate anus and anal stenosis have standard surgical treatment, and renal and genitourinary anomalies are followed in nephrology and urology [1]. The sources consulted do not include the classification of laryngeal clefts or specific surgical protocols.

7.8. Development, behaviour and agents to use with caution

Children with delay benefit from early intervention and individualised educational support [1, 2, 7]. Behavioural manifestations are treated in psychology or psychiatry [1]. Stimulants used in ADHD should be considered carefully, because the hamartoma is a central nervous system lesion that predisposes to seizures, and seizures may be triggered or worsened by these drugs [1]. GnRH agonists carry warnings of seizures and psychiatric events [156, 158]. Underdosing or abrupt discontinuation of hydrocortisone can precipitate adrenal crisis [133].

Table 7.15 — Interventions to avoid or to use with caution

7.9. Pregnancy, perioperative care and transition

There are no pregnancy-specific guidelines for PHS. In women with gelastic epilepsy, antiseizure medications reduce the risk of maternal mortality associated with seizures; fetal exposure may increase the risk of malformations, depending on the drug, dose and stage of gestation, but this risk is generally lower than that of untreated maternal seizures, and the discussion should take place before conception, with a switch to the lowest-risk drug when possible [1]. Pregnancy with hypopituitarism follows the general guidelines, summarised in section 7.2.4 [1, 117, 151]. In active labour, the Endocrine Society guideline recommends the hydrocortisone stress dose used in major surgery [115]. GnRH agonists are contraindicated in pregnancy [156].

In the perioperative care of PHS with adrenal insufficiency, the stress doses in tables 7.2 and 7.3 apply to all surgery, including correction of polydactyly and anorectal surgery [115]. Prior laryngoscopy identifies bifid epiglottis and laryngeal clefts before intubation [1]. At the transition to adulthood, counselling about the 50% risk to offspring and pregnancy planning should be offered to affected young adults [1, 2].

7.10. Clinical trials and developments in the last 12 months

On ClinicalTrials.gov, consulted on 17 October 2026, there are no interventional studies directed at PHS, GCPS or GLI3; the only PHS record is the observational study of the National Human Genome Research Institute NCT00001404 ("Phenotype and Etiology of Pallister-Hall Syndrome"), started on 18 August 1994, with 1170 participants, completed [180]. The feasibility and safety trial NCT03028246 (InSightec), of MRI-guided focused ultrasound in central benign intracranial tumours in paediatric and young adult patients, was recruiting, with 20 planned participants and primary completion expected on 31 December 2026, with no results submitted [177]. The registered trials of indirect relevance (table 7.16) are studies of new GnRH agonist formulations in central precocious puberty of all causes and observational cohort studies with hamartoma or rare syndromes; no hamartoma-specific LITT, RFTC or radiosurgery trial was identified other than NCT03028246, and the European trials register (CTIS) was not consulted [181–184]. A negative search does not prove that other studies do not exist.

Table 7.16 — Studies registered on ClinicalTrials.gov relevant to GLI3 disorders (consulted 17-10-2026)

Table 7.17 — Developments between October 2025 and October 2026, by level of evidence

The 2021 review mentions preclinical studies with CB2 cannabinoid receptor agonists in hamartoma slices (unpublished data) and SMO inhibitors as a therapeutic hypothesis, without a clinical trial [17]. No molecular therapy in clinical development for GLI3 disorders was identified.

7.11. Recommendations of the sources and divergences

Table 7.18 — Divergences between sources

8. Reproductive Options and Family Counselling

8.1. Recurrence risks by scenario

Table 8.1 — Recurrence risks

About 75% of patients with PHS have an affected parent and about 25% have a de novo variant; de novo cases are generally more severe than familial ones [1]. In the French series, de novo origin was confirmed in all 13 sporadic PHS cases studied and in 6 of the 16 sporadic GCPS cases [51]. In GCPS, the proportion of de novo cases is unknown, because subtle signs in the parents are poorly assessed and molecular data are insufficient [2]. An apparently isolated case may result from failure to recognise the disorder in relatives or from reduced penetrance, and only molecular testing of both parents confirms a de novo variant [2]. The risk to other relatives depends on the status of the parents: if a parent is affected, has the variant or, in GCPS, a balanced structural rearrangement, relatives on that side may be at risk [1, 2]. Severity in affected siblings tends to be similar to that of the proband, because intrafamilial variability is lower than interfamilial variability, but a mosaic parent may have minimal signs and more severely affected offspring [1, 2, 7]. In non-syndromic hamartoma, the presence of a germline variant in a ciliary gene, with a second somatic variant, alters the recurrence risk, and the estimate of non-increased risk applies only to an exclusively somatic variant [75, 78].

8.2. Assessment of relatives and cascade testing

In a simplex case, the parents should be examined and tested for the proband's variant [1, 2]. In GCPS, examination of the parents includes radiographs of the hands and feet and, if there are signs of hydrocephalus, neuroimaging; if the proband has a 7p14.1 deletion, chromosome analysis should be offered to the parents to exclude a balanced rearrangement [2]. In PHS, relatives may undergo testing for the variant or, without testing, examination for polydactyly, laryngoscopy (bifid epiglottis) and MRI (hamartoma); cortisol assessment is urgent in newborns of families with PHS and cortisol deficiency [1]. The GCPS GeneReviews does not recommend a specific clinical assessment of at-risk relatives beyond counselling, unlike the PHS GeneReviews, in which early testing has therapeutic consequences [1, 2]. Smith's emphasises that complete assessment of the parents, including brain MRI in some cases, is needed to estimate the recurrence risk [42]. In the syndactylies, the Oxford Desk Reference recommends explaining the minimal carrier signs to the parents, with a pedigree of at least three generations [58]. In the family of Ng and colleagues, the parents had no extra digits, bifid epiglottis or hamartoma, which shows that a normal examination does not exclude germline mosaicism [79].

8.3. Prenatal and preimplantation diagnosis

Prenatal testing by amniocentesis or chorionic villus sampling and PGT are possible when the familial variant, or the parental rearrangement, is known [1, 2, 7]. In PHS, prenatal diagnosis allows immediate treatment of adrenal insufficiency after birth to be prepared [1]. High-resolution ultrasound detects polydactyly, but its absence does not exclude the disorder [1, 2, 7]. The parents' preference for prenatal diagnosis of a generally mild disorder, such as typical GCPS, is a matter of informed decision; the GeneReviews chapters note differences in perspective among professionals and within families and consider it a personal decision that discussion can help with, and the sources consulted do not describe recommendations on termination of pregnancy [1, 2]. In the French series, the seven pregnancies with a fetus with severe PHS were terminated under French law [51].

8.4. Counselling

Preconception counselling should be offered to young adults who are affected or at risk of being carriers [1, 2]. In minors, testing of asymptomatic relatives has direct clinical benefit in PHS (assessment of cortisol and the larynx) [1]. For families with PHS, GeneReviews lists the resources of MedlinePlus (entries "Pallister-Hall syndrome" and "Polydactyly"), the American Epilepsy Society and the Epilepsy Foundation (epilepsy.com; telephone 800-332-1000), whose "Toolbox" supports non-medical interventions and coping strategies; for GCPS, those of MedlinePlus, the Children's Craniofacial Association (ccakids.org; telephone 800-535-3643) and Face Equality International [1, 2]. The author of the PHS GeneReviews asks clinicians with families with atypical manifestations or without an identified variant to make contact [1]. The Oxford Desk Reference lists Contact a Family and the National Organization for Rare Disorders (NORD) for families with polydactyly [13].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each number measures

Table 9.1 — Epidemiological estimates

The Smith's figure of 1–9/1,000,000 for the incidence of GCPS has no primary source in the book and coincides with the expert estimate of the 2008 review [7, 38]. The sources consulted do not include population-based prevalence studies of PHS or GCPS in any country.

9.2. United States, Europe and populations with consanguinity

Most of the clinical and molecular data come from the NIH cohort in the USA (observational study NCT00001404, started in 1994) [24, 25, 180]. In Europe, the data come from cohorts in France (76 cases from 55 families, with cases from Tunisia, Lebanon, India and South Africa), the Netherlands (297 cases, including published cases, with 46 patients from the Sophia Children's Hospital in Rotterdam), Germany (48 cases from 16 families), Poland (16 probands), Belgium (13 patients), Austria (5 patients with a 7p deletion), Lithuania (three-generation family) and Spain (prenatal diagnosis) [21, 26, 51–53, 100, 102, 109]. The series of somatic hypothalamic hamartoma include a French series and an international series [75, 76]. The sources do not describe GLI3 founder variants. Consanguinity has two implications: the recessive forms of syndromic polydactyly due to homozygous GLI3 variants (three families) and the risk of a homozygous fetus in couples with PAPA1/PAPB, described in Iran [72, 74]. Cases have been published in Asia (Japan, China), the Middle East (Saudi Arabia) and Latin America (Colombia, first case in the country) [61, 81, 82, 85].

9.3. African ancestry

Isolated postaxial polydactyly is about 10 times more common in people of African ancestry [13]. Therefore, PAP-B should be assessed with caution as a criterion for GCPS or PHS in people from West and Central Africa, and in the assessment of relatives of patients with PHS it counts only in relatives who are not of that ancestry [1, 2].

9.4. Portugal and Brazil

In Portugal, Raposo and colleagues published in 2015 a prenatal diagnosis of GCPS (Centro Hospitalar Tondela-Viseu), which is the reference used by Orphanet for the European prevalence class [6, 110]. In Brazil, Castro and colleagues, in Rio de Janeiro, described the prenatal diagnosis of PHS by ultrasound, MRI and three-dimensional reconstruction, and a 2026 imaging review of PHS has a Brazilian first author [111, 186]. The Brazilian book of genetic syndromes includes PHS in the differential of the VACTERL association and among the syndromes with holoprosencephaly and a normal karyotype (about 25% of cases of holoprosencephaly, according to OMIM as cited by the book) [122, 187]. The 2026 international consensus on hamartoma included a Brazilian centre (São Paulo Epilepsy Clinic), the only one from South America [28]. The sources consulted do not include population data from Portugal or Brazil or Portuguese or Brazilian patient associations specific to these disorders, which does not demonstrate that they do not exist.

9.5. Special populations

In adults with late-diagnosed PHS, hormone deficiencies and malformations not recognised in childhood should be sought [81, 82]. In children with intellectual disability or ASD, recognition of gelastic seizures requires video and video-EEG [28]. In pregnancy, the considerations in section 7.9 apply.

9.6. European Reference Networks and patient associations

The relevant ERNs are ITHACA (rare malformation syndromes and intellectual disability), ERN BOND (bone and limb disorders), ERN CRANIO (craniofacial and ear, nose and throat anomalies), Endo-ERN (rare endocrine disorders, including hypothalamus-pituitary and puberty), EpiCARE (rare and complex epilepsies), ERNICA (digestive and anorectal malformations) and ERKNet (kidney diseases, including CAKUT) [30–36]. The Freiburg, Paris (Fondation Rothschild), Rome (Bambino Gesù) and Lyon centres that took part in the 2026 consensus are members of EpiCARE, and Great Ormond Street Hospital (London) is an affiliated partner centre; the management of the network is funded by the European Commission [28, 32]. The Hope for Hypothalamic Hamartomas association maintains a patient registry, a medical board and research funding, and organises the international hypothalamic hamartoma symposia (Washington in 2019, Calgary in 2022, Barcelona in 2025), whose medical board initiated the 2026 consensus [28, 37]. For septo-optic dysplasia, an endocrine differential, the Oxford Desk Reference lists the FOCUS Families association [103].

10. Illustrative Clinical Case

10.1. Pallister-Hall syndrome diagnosed in adulthood

Mukhtar and colleagues published in 2026 the case of a 26-year-old Saudi man, born to non-consanguineous parents, with arterial hypertension treated with amlodipine 5 mg/day for four years and recurrent hypoglycaemia for one year (two to four episodes per month), fasting and postprandial, with capillary blood glucose between 2.7 and 3.3 mmol/L and, in some episodes, loss of consciousness lasting less than one minute [81]. His mother, a sister, his maternal grandmother and several maternal uncles had polydactyly [81]. On examination he had class III obesity (179 cm, 136 kg, body mass index of 42.4 kg/m²), blood pressure of 146/88 mmHg, mild acanthosis nigricans and polydactyly of the right hand; an extra toe had been removed in childhood [81].

Hormonal assessment showed thyroid-stimulating hormone (TSH) of 3.1 mU/L, free thyroxine of 16.3 pmol/L, morning cortisol of 493 nmol/L, ACTH of 42 ng/L, insulin-like growth factor 1 (IGF-1) of 152 ng/mL, testosterone of 10.9 nmol/L, prolactin of 20.9 µg/L (slightly elevated), parathyroid hormone of 64.6 ng/L with 25-hydroxyvitamin D of 20 nmol/L and glycated haemoglobin of 4.9%; LH, FSH and dynamic tests were not measured [81]. In the 72-hour fast, glucose fell to 2.4 mmol/L at 68 hours, with non-suppressed insulin of 73 pmol/L and C-peptide of 0.72 nmol/L, and recovered spontaneously; in the mixed-meal test (45 g of carbohydrate), the nadir was 2.8 mmol/L with insulin of 163 pmol/L and C-peptide of 2.9 nmol/L, and peak insulin reached 3135 pmol/L [81]. Screening for secretagogues was negative, and abdominal MRI showed no pancreatic lesion, but did show an atrophic, scarred right kidney and hepatic steatosis [81].

Exome sequencing identified the heterozygous variant NM_000168.6:c.2059del p.(Glu687LysfsTer6) in exon 12 of GLI3 (rs2128709905; ClinVar VCV 1120244, pathogenic), confirmed by Sanger sequencing [63, 81]. The variant lies in the middle third of the gene, in the region associated with PHS [25]. Only after the molecular diagnosis was the targeted assessment carried out: brain MRI showed a hypothalamic lesion of about 12 × 12 × 7.5 mm, probably originating from the tuber cinereum, involving both mammillary bodies, isointense to cortex on T1 and T2 and without contrast enhancement, consistent with hamartoma, with a slightly thickened and deviated pituitary stalk; laryngoscopy showed a partially bifid epiglottis; radiography of the hand showed fusion of the right 3rd and 4th metacarpals with an insertional digit at the base of the 4th (mesoaxial polydactyly), and radiography of the foot showed previous resection of a postaxial toe [81]. The patient thus met the clinical criterion for PHS (hamartoma and mesoaxial polydactyly) [1].

Routine EEG and 24-hour video-EEG were normal, no antiseizure medication was started and the patient was referred to neurosurgery and genetics, with dietary measures for the hypoglycaemia (small, frequent meals, avoiding simple carbohydrates); at six months he still had episodes, without further intervention [81]. The authors acknowledge that they did not completely exclude a small insulinoma, autoimmune hypoglycaemia and other causes, that they considered the hamartoma a diagnostic feature and not the demonstrated cause of the hypoglycaemia, and that the relatives were not tested [81]. Neurosurgical referral for an asymptomatic hamartoma in a patient with PHS diverges from the GeneReviews recommendation not to resect the hamartoma in PHS outside very unusual circumstances and from the 2026 consensus, which reserves surgery for epilepsy refractory to two antiseizure medications [1, 28]. Figure 1 of the article shows the hands, the radiographs of the hand and foot and the bifid epiglottis (https://pmc.ncbi.nlm.nih.gov/articles/PMC13649492/figure/f1/), figure 2 the chromatogram of the deletion (https://pmc.ncbi.nlm.nih.gov/articles/PMC13649492/figure/f2/) and figure 3 the MRI of the hamartoma on sagittal and coronal T2 (https://pmc.ncbi.nlm.nih.gov/articles/PMC13649492/figure/f3/); the figures were consulted through the legends of the full text [81].

The case illustrates three points: familial PHS may go unnoticed until adulthood when polydactyly is the only recognised sign; bifid epiglottis and hamartoma are found only when they are sought; and a family with several people with polydactyly should raise the possibility of mild PHS and not only isolated PAP-A [1, 81].

10.2. Complementary case: Greig syndrome in three generations

Siavrienė and colleagues described a non-consanguineous Lithuanian family [21]. The proband, aged 6 years, was born at 39 weeks weighing 3500 g with an Apgar score of 9/10, after a threatened miscarriage at 7 weeks; at birth she had a pedunculated accessory digit on the ulnar border of the 5th finger of both hands (PAP-B), asymmetric cutaneous syndactyly (partial from the 2nd to the 4th fingers on the right and complete from the 2nd to the 5th on the left), preaxial polydactyly of the feet and complete, symmetric syndactyly of the first three toes; she underwent hand surgery between 1 and 4 years of age; at 6 years she had a head circumference of 55 cm (above the 97th centile), macrocephaly and broad thumbs [21]. The mother (4600 g at birth) had bilateral PAP-B, syndactyly, preaxial polydactyly of the feet, mild hydrocephalus on CT at 27 years, a head circumference of 61 cm at 32 years and iatrogenic pain after foot surgery; the grandmother had polysyndactyly of the hands and feet, camptodactyly and broad thumbs, with a head circumference of 57 cm (75th centile) at 56 years [21]. All three had the novel variant NM_000168.6:c.473+3A>T, absent from population databases; RNA study of the grandmother's fibroblasts showed skipping of exon 4, with predicted p.(His123Argfs*57), whereas no GLI3 RNA was detected in blood; the missense variant c.2179G>A p.(Gly727Arg), inherited from the proband's healthy father, did not segregate with the disease [21]. The congenital chorioretinitis, macular dystrophy and other ocular findings in the family are not attributed to GLI3 [21]. Figure 1 of the article shows the pedigree, the hands and feet and the radiograph of the proband's feet before surgery, at 1 year (https://pmc.ncbi.nlm.nih.gov/articles/PMC6732282/figure/mgg3878-fig-0001/), consulted through the legend [21]. The family shows the variability of head circumference across generations and the need to study RNA in a tissue that expresses the gene.

10.3. Complementary case: GCPS due to an 18 Mb contiguous gene deletion

Kozma and colleagues described a Romanian boy referred at 3 years of age to the genetics department of the "Dr. Gavril Curteanu" Municipal Clinical Hospital (Oradea) for psychomotor delay and multiple congenital anomalies, the second child of healthy unrelated parents [9]. He was born by vaginal delivery at 34 weeks weighing 3030 g (97th centile), 48 cm long (90th centile) and with a head circumference of 36 cm (1 cm above the 97th centile), with Apgar scores of 5/7/8 and suspected perinatal hypoxia [9]. He had macro-acrocephaly (head circumference of 53 cm at 3 years), a prominent metopic suture, a high hairline, hypertelorism, telecanthus, upslanting palpebral fissures, convergent strabismus, a broad, depressed nasal root, anteverted nares, a long philtrum, microdontia and low-set ears; broad, square thumbs, a scar from correction of preaxial polydactyly, a broad hallux with complete cutaneous syndactyly between the 1st and 2nd and between the 2nd and 3rd toes on both feet, flat feet and joint laxity; congenital stridor, umbilical hernia and right cryptorchidism operated on at 16 months [9]. Development was very delayed (head control at 6–7 months, sat at 2 years, no independent walking, few syllables; mental age of 8 months on the Bayley II scale at 3 years and of 15 months at 6 years), and MRI at 1 and 3 years showed hypoplasia of the corpus callosum, aqueductal stenosis and mild supratentorial ventriculomegaly [9]. The karyotype was 46,XY,del(7)(p13p15), and the 180K microarray showed an 18.37 Mb interstitial deletion at 7p14.2-p11.2 (arr[hg38](35830920_54201451)del), with 205 genes, 22 of which were OMIM genes with a phenotype, among them GLI3, GCK (MODY2), ANLN (kidney disease), POU6F2 (Wilms tumour), CCM2 (cavernomas), IKZF1 (immunodeficiency) and CDK13, CAMK2B and RALA (intellectual disability), which justifies surveillance targeted at these comorbidities [9]. The parents declined testing, so de novo origin was presumed without excluding a balanced parental rearrangement [9]. At 6 years, growth had normalised (22 kg, 116 cm, head circumference of 53 cm), with slight psychomotor improvement; treatment consisted of surgical correction of the digits and cognitive and motor stimulation [9]. Figure 1 of the article shows the face at 1 and 3 years and the hands and feet (https://pmc.ncbi.nlm.nih.gov/articles/PMC8623992/figure/genes-12-01674-f001/), consulted through the legend of the full text [9]. The case illustrates the phenotype of GCPS-CGS, with developmental impairment and brain anomalies, the need for a microarray that delimits the deletion and the neighbouring genes, and the importance of parental testing for counselling, which here remained incomplete.

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

In typical GCPS, the prognosis is good, with few medical complications and normal life expectancy; several large families with a mild form have good general health and normal longevity [2, 7]. The cognitive risk in typical GCPS has not been shown to exceed the baseline risk of 2–3% and is concentrated in GCPS-CGS, which has frequent cognitive impairment and seizures [7, 56]. In the Leuven series, intellectual disability occurred in 3 of 13 patients, all with a large deletion [102].

In PHS, the prognosis of cases without a family history depends on the malformations present, and literature reviews are biased towards severity [1]. Early lethality results mainly from panhypopituitarism, severe airway malformations (laryngotracheal clefts) and unrecognised imperforate anus; in the absence of potentially fatal malformations, the prognosis should be presumed good in non-familial cases and, in families, follows familial severity [1]. Smith's states that death before 3 years of age "is not uncommon", that several affected individuals have had offspring and that normal intellectual ability has been observed [42]. Fetuses with severe malformations (agnathia, limb reduction) represent the lethal prenatal extreme [51]. The family with 22 affected members of Kang and colleagues had low morbidity [48].

In hypothalamic hamartoma of all causes, seizure freedom is 50% after the first procedure and 64.5% after multiple treatments, and cognitive and psychiatric comorbidities may persist [107, 178]. Generalised tonic-clonic seizures, a risk factor for sudden unexpected death in epilepsy, are relatively rare in hamartoma [17]. The 2026 consensus considers hamartoma surgery safe, with limited morbidity and no significant risk of mortality, and associates early surgical assessment with better outcomes [28]. In PHS, seizures were generally well controlled with antiseizure medications in the NIH series, unlike isolated hamartoma [90].

11.2. Surveillance

Table 11.1 — Recommended surveillance

12. Summary and Final Considerations

GLI3-related disorders form an allelic spectrum in which the type and position of the variant predict the phenotype. Haploinsufficiency, caused by variants of all classes and by deletions, produces Greig cephalopolysyndactyly syndrome and the apparently isolated polysyndactylies; truncating variants in the middle third of the gene, which escape mRNA decay and generate a constitutive repressor, produce Pallister-Hall syndrome; large 7p14 deletions add the phenotype of the neighbouring genes; and somatic variants restricted to the hamartoma explain some non-syndromic hypothalamic hamartomas. The variant c.2374C>T p.(Arg792Ter), in the PHS region but associated with GCPS, shows the limits of the positional rule.

GCPS is, in most patients, a disorder of the limbs and craniofacies with a good prognosis, and its treatment is elective surgery for polydactyly, with priority to preaxial polydactyly of the hands and caution in the feet. Intellectual disability, seizures and diabetes due to loss of GCK suggest a contiguous gene deletion, which chromosomal microarray delimits. The distinction between GCPS due to deletion and KIF7 acrocallosal syndrome changes the recurrence risk for the parents.

PHS is more often mild than severe, and mild forms are diagnosed as isolated postaxial polydactyly if the hamartoma and bifid epiglottis are not sought. The risk to life is concentrated in the neonatal period: cortisol deficiency, laryngotracheal clefts and unrecognised imperforate anus. Assessment of cortisol before any other pituitary axis and stress-dose hydrocortisone without waiting for results are the measures the sources identify as determining neonatal survival. Replacement of the other axes follows the general guidelines, with levothyroxine started only once the corticotroph axis is secured, titration by FT4 and not by TSH, reassessment of the thyroid and adrenal axes when GH is started and attention to interactions between enzyme-inducing antiseizure medications and hydrocortisone, levothyroxine and desmopressin. The PHS hamartoma is not resected routinely, and seizures generally respond to antiseizure medications; surgery, preferably by minimally invasive ablative techniques, is reserved for epilepsy refractory to two antiseizure medications, according to the 2026 international consensus. Central precocious puberty is treated with GnRH agonists according to the 2026 Endocrine Society guideline, which diverges from the US labels on biochemical monitoring and whose annual implant is not available in Europe.

Molecular diagnosis detects a variant in about 95% of patients with clinical PHS (19/20 and 20/22 in the NIH series) and in 68% of probands with typical GCPS, with a breakdown estimated by GeneReviews at about 80% sequence variants and 20% deletions; the limits of the PHS region, c.1998 to c.3481, are valid on the MANE transcript NM_000168.6; a negative result does not exclude the diagnosis, and novel truncating variants in the PHS region require specialised interpretation, because the PVS1 criterion does not apply automatically to a mechanism that does not depend on mRNA decay. Counselling distinguishes autosomal dominant inheritance with a 50% risk, germline mosaicism in de novo cases, balanced parental rearrangements in deletions, recessive forms due to SMO or to GLI3 homozygosity with consanguinity, and somatic hamartoma without an increased recurrence risk. There are no population-based prevalence studies, and the available figures are expert classes and counts of published cases.

13. References

1. Biesecker LG. GLI3-Related Pallister-Hall Syndrome. 2000 May 25 [Updated 2022 Aug 18; Revised 2024 Feb 22]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1465/
2. Biesecker LG, Johnston JJ. GLI3-Related Greig Cephalopolysyndactyly Syndrome. 2001 Jul 9 [Updated 2025 Aug 28]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1446/
3. OMIM. 146510 PALLISTER-HALL SYNDROME; PHS. Online Mendelian Inheritance in Man. Johns Hopkins University. Available from: https://omim.org/entry/146510. Accessed 17 Oct 2026.
4. Orphanet. Pallister-Hall syndrome (ORPHA:672). Available from: https://www.orpha.net/en/disease/detail/672. Accessed 17 Oct 2026.
5. OMIM. 175700 GREIG CEPHALOPOLYSYNDACTYLY SYNDROME; GCPS. Online Mendelian Inheritance in Man. Johns Hopkins University. Available from: https://omim.org/entry/175700. Accessed 17 Oct 2026.
6. Orphanet. Greig cephalopolysyndactyly syndrome (ORPHA:380). Available from: https://www.orpha.net/en/disease/detail/380. Accessed 17 Oct 2026.
7. Biesecker LG. The Greig cephalopolysyndactyly syndrome. Orphanet J Rare Dis. 2008;3:10. doi:10.1186/1750-1172-3-10. PMID: 18435847. PMC2397380.
8. Orphanet. Greig cephalopolysyndactyly-contiguous gene syndrome (ORPHA:658805). Available from: https://www.orpha.net/en/disease/detail/658805. Accessed 17 Oct 2026.
9. Kozma K, Bembea M, Jurca CM, Ioana M, Streață I, Şoşoi SŞ, et al. Greig Cephalopolysyndactyly Contiguous Gene Syndrome: Case Report and Literature Review. Genes (Basel). 2021;12(11):1674. doi:10.3390/genes12111674. PMID: 34828280. PMC8623992.
10. OMIM. 165240 GLI-KRUPPEL FAMILY MEMBER 3; GLI3. Online Mendelian Inheritance in Man. Johns Hopkins University. Available from: https://omim.org/entry/165240. Accessed 17 Oct 2026.
11. Orphanet. Postaxial polydactyly type A (ORPHA:93334). Available from: https://www.orpha.net/en/disease/detail/93334. Accessed 17 Oct 2026.
12. Orphanet. Postaxial polydactyly type B (ORPHA:93335). Available from: https://www.orpha.net/en/disease/detail/93335. Accessed 17 Oct 2026.
13. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 284–287.
14. Ortiz-Cruz G, Luna-Muñoz L, Arteaga-Vázquez J, Mutchinick OM. Isolated postaxial polydactyly: Epidemiologic characteristics from a multicenter birth defects study. Am J Med Genet A. 2019;179(8):1432-1441. doi:10.1002/ajmg.a.61193. PMID: 31091006.
15. Orphanet. Polysyndactyly (ORPHA:93338). Available from: https://www.orpha.net/en/disease/detail/93338. Accessed 17 Oct 2026.
16. Orphanet. Gelastic seizures with hypothalamic hamartoma (ORPHA:86906). Available from: https://www.orpha.net/en/disease/detail/86906. Accessed 17 Oct 2026.
17. Cohen NT, Cross JH, Arzimanoglou A, Berkovic SF, Kerrigan JF, Miller IP, et al. Hypothalamic Hamartomas: Evolving Understanding and Management. Neurology. 2021;97(18):864-873. doi:10.1212/WNL.0000000000012773. PMID: 34607926. PMC8610628.
18. Orphanet. Acrocallosal syndrome (ORPHA:36). Available from: https://www.orpha.net/en/disease/detail/36. Accessed 17 Oct 2026.
19. Orphanet. Isolated tibial hemimelia (ORPHA:93322). Available from: https://www.orpha.net/en/disease/detail/93322. Accessed 17 Oct 2026.
20. OMIM. 241800 PALLISTER-HALL-LIKE SYNDROME; PHLS. Online Mendelian Inheritance in Man. Johns Hopkins University. Available from: https://omim.org/entry/241800. Accessed 17 Oct 2026.
21. Siavrienė E, Mikštienė V, Radzevičius D, Maldžienė Ž, Rančelis T, Petraitytė G, et al. Novel GLI3 variant causes Greig cephalopolysyndactyly syndrome in three generations of a Lithuanian family. Mol Genet Genomic Med. 2019;7(9):e878. doi:10.1002/mgg3.878. PMID: 31325247. PMC6732282.
22. Blake J, Hu D, Cain JE, Rosenblum ND. Urogenital development in Pallister-Hall syndrome is disrupted in a cell-lineage-specific manner by constitutive expression of GLI3 repressor. Hum Mol Genet. 2016;25(3):437-47. doi:10.1093/hmg/ddv483. PMID: 26604140. PMC4731018.
23. GeneGuidance. GLI3 — gene and condition curation [gene–disease curation database, version of 2026-09-26].
24. Johnston JJ, Sapp JC, Turner JT, Amor D, Aftimos S, Aleck KA, et al. Molecular analysis expands the spectrum of phenotypes associated with GLI3 mutations. Hum Mutat. 2010;31(10):1142-54. doi:10.1002/humu.21328. PMID: 20672375. PMC2947617.
25. Johnston JJ, Olivos-Glander I, Killoran C, Elson E, Turner JT, Peters KF, et al. Molecular and clinical analyses of Greig cephalopolysyndactyly and Pallister-Hall syndromes: robust phenotype prediction from the type and position of GLI3 mutations. Am J Hum Genet. 2005;76(4):609-22. doi:10.1086/429346. PMID: 15739154. PMC1199298.
26. Schwarzbraun T, Windpassinger C, Ofner L, Vincent JB, Cheung J, Scherer SW, et al. Genomic analysis of five chromosome 7p deletion patients with Greig cephalopolysyndactyly syndrome (GCPS). Eur J Med Genet. 2006;49(4):338-45. doi:10.1016/j.ejmg.2005.10.133. PMID: 16829355.
27. Persani L, Brabant G, Dattani M, Bonomi M, Feldt-Rasmussen U, Fliers E, et al. 2018 European Thyroid Association (ETA) Guidelines on the Diagnosis and Management of Central Hypothyroidism. Eur Thyroid J. 2018;7(5):225-37. doi:10.1159/000491388. PMID: 30374425. PMC6198777.
28. Cohen NT, Li X, Berl MM, Oluigbo CO, Shirozu H, Berkovic SF, et al. International Consensus on the Evaluation and Management of Hypothalamic Hamartomas: Results From a Modified Delphi Survey. Neurology. 2026;106(7):e214760. doi:10.1212/WNL.0000000000214760. PMID: 41818657. PMC13031823.
29. Latronico AC, Roberts SA, Alonzo M, Argente J, Canton APM, Carel JC, et al. Central precocious puberty: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2026;111(8):2101-2144. doi:10.1210/clinem/dgag168. PMID: 42287186.
30. ERN ITHACA — European Reference Network for Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders. Available from: https://ern-ithaca.eu/
31. Endo-ERN — European Reference Network on Rare Endocrine Conditions. Available from: https://endo-ern.eu/
32. ERN EpiCARE — European Reference Network for Rare and Complex Epilepsies. Available from: https://epi-care.eu/
33. ERNICA — European Reference Network for rare Inherited and Congenital digestive and gastrointestinal Anomalies. Available from: https://www.ern-ernica.eu/
34. ERKNet — European Rare Kidney Disease Reference Network. Available from: https://www.erknet.org/
35. ERN CRANIO — European Reference Network for craniofacial anomalies and ENT disorders. Available from: https://www.ern-cranio.eu/
36. ERN BOND — European Reference Network on Rare Bone Diseases. Available from: https://ernbond.eu/
37. Hope for Hypothalamic Hamartomas. Available from: https://hopeforhh.org/
38. Jones KL, Jones MC, del Campo M. Greig cephalopolysyndactyly syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 594–595.
39. Vortkamp A, Gessler M, Grzeschik KH. GLI3 zinc-finger gene interrupted by translocations in Greig syndrome families. Nature. 1991;352(6335):539-40. doi:10.1038/352539a0. PMID: 1650914.
40. Kalff-Suske M, Wild A, Topp J, Wessling M, Jacobsen EM, Bornholdt D, et al. Point mutations throughout the GLI3 gene cause Greig cephalopolysyndactyly syndrome. Hum Mol Genet. 1999;8(9):1769-77. doi:10.1093/hmg/8.9.1769. PMID: 10441342.
41. Hall JG, Pallister PD, Clarren SK, Beckwith JB, Wiglesworth FW, Fraser FC, et al. Congenital hypothalamic hamartoblastoma, hypopituitarism, imperforate anus and postaxial polydactyly--a new syndrome? Part I: clinical, causal, and pathogenetic considerations. Am J Med Genet. 1980;7(1):47-74. doi:10.1002/ajmg.1320070110. PMID: 7211952.
42. Jones KL, Jones MC, del Campo M. Pallister-Hall syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 248–249.
43. Narumi Y, Kosho T, Tsuruta G, Shiohara M, Shimazaki E, Mori T, et al. Genital abnormalities in Pallister-Hall syndrome: Report of two patients and review of the literature. Am J Med Genet A. 2010;152A(12):3143-7. doi:10.1002/ajmg.a.33720. PMID: 21108399.
44. Hall JG. Pallister-Hall syndrome has gone the way of modern medical genetics. Am J Med Genet C Semin Med Genet. 2014;166C(4):414-8. doi:10.1002/ajmg.c.31419. PMID: 25424727.
45. Hall JG. The early history of Pallister-Hall syndrome-Buried treasure of a sort. Gene. 2016;589(2):100-3. doi:10.1016/j.gene.2016.01.003. PMID: 26768579.
46. Biesecker LG, Abbott M, Allen J, Clericuzio C, Feuillan P, Graham JM Jr, et al. Report from the workshop on Pallister-Hall syndrome and related phenotypes. Am J Med Genet. 1996;65(1):76-81. doi:10.1002/(SICI)1096-8628(19961002)65:1<76::AID-AJMG12>3.0.CO;2-O. PMID: 8914745.
47. Biesecker LG, Graham JM Jr. Pallister-Hall syndrome. J Med Genet. 1996;33(7):585-9. doi:10.1136/jmg.33.7.585. PMID: 8818945. PMC1050667.
48. Kang S, Allen J, Graham JM Jr, Grebe T, Clericuzio C, Patronas N, et al. Linkage mapping and phenotypic analysis of autosomal dominant Pallister-Hall syndrome. J Med Genet. 1997;34(6):441-6. doi:10.1136/jmg.34.6.441. PMID: 9192261. PMC1050964.
49. Kang S, Graham JM Jr, Olney AH, Biesecker LG. GLI3 frameshift mutations cause autosomal dominant Pallister-Hall syndrome. Nat Genet. 1997;15(3):266-8. doi:10.1038/ng0397-266. PMID: 9054938.
50. Wallace RH, Freeman JL, Shouri MR, Izzillo PA, Rosenfeld JV, Mulley JC, et al. Somatic mutations in GLI3 can cause hypothalamic hamartoma and gelastic seizures. Neurology. 2008;70(8):653-5. doi:10.1212/01.wnl.0000284607.12906.c5. PMID: 18057317.
51. Démurger F, Ichkou A, Mougou-Zerelli S, Le Merrer M, Goudefroye G, Delezoide AL, et al. New insights into genotype-phenotype correlation for GLI3 mutations. Eur J Hum Genet. 2015;23(1):92-102. doi:10.1038/ejhg.2014.62. PMID: 24736735. PMC4266745.
52. Baas M, Burger EB, van den Ouweland AM, Hovius SE, de Klein A, van Nieuwenhoven CA, et al. Variant type and position predict two distinct limb phenotypes in patients with GLI3-mediated polydactyly syndromes. J Med Genet. 2021;58(6):362-368. doi:10.1136/jmedgenet-2020-106948. PMID: 32591344. PMC8142428.
53. Sczakiel HL, Hülsemann W, Holtgrewe M, Abad-Perez AT, Elsner J, Schwartzmann S, et al. GLI3 variants causing isolated polysyndactyly are not restricted to the protein's C-terminal third. Clin Genet. 2021;100(6):758-765. doi:10.1111/cge.14059. PMID: 34482537.
54. Al-Qattan MM, Shamseldin HE, Salih MA, Alkuraya FS. GLI3-related polydactyly: a review. Clin Genet. 2017;92(5):457-466. doi:10.1111/cge.12952. PMID: 28224613.
55. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 848–850.
56. Johnston JJ, Olivos-Glander I, Turner J, Aleck K, Bird LM, Mehta L, et al. Clinical and molecular delineation of the Greig cephalopolysyndactyly contiguous gene deletion syndrome and its distinction from acrocallosal syndrome. Am J Med Genet A. 2003;123A(3):236-42. doi:10.1002/ajmg.a.20318. PMID: 14608643.
57. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166. Tool: https://variantvalidator.org/ (GRCh38, MANE Select transcripts)
58. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 328–329.
59. Saitsu H, Sonoda M, Higashijima T, Shirozu H, Masuda H, Tohyama J, et al. Somatic mutations in GLI3 and OFD1 involved in sonic hedgehog signaling cause hypothalamic hamartoma. Ann Clin Transl Neurol. 2016;3(5):356-65. doi:10.1002/acn3.300. PMID: 27231705. PMC4863748.
60. Pagnamenta AT, Yu J, Evans J, Twiss P, Genomics England Research Consortium, Musculoskeletal GeCIP MDT, et al. Conclusion of diagnostic odysseys due to inversions disrupting GLI3 and FBN1. J Med Genet. 2023;60(5):505-510. doi:10.1136/jmg-2022-108753. PMID: 36411030. PMC10176330.
61. Ito S, Kitazawa R, Haraguchi R, Kondo T, Ouchi A, Ueda Y, et al. Novel GLI3 variant causing overlapped Greig cephalopolysyndactyly syndrome (GCPS) and Pallister-Hall syndrome (PHS) phenotype with agenesis of gallbladder and pancreas. Diagn Pathol. 2018;13(1):1. doi:10.1186/s13000-017-0682-8. PMID: 29368652. PMC6389258.
62. Abdullah, Yousaf M, Azeem Z, Bilal M, Liaqat K, Hussain S, et al. Variants in GLI3 Cause Greig Cephalopolysyndactyly Syndrome. Genet Test Mol Biomarkers. 2019;23(10):744-50. doi:10.1089/gtmb.2019.0071. PMID: 31573334.
63. National Center for Biotechnology Information. ClinVar: GLI3 variation records VCV000013820, VCV000013821, VCV000013825, VCV000013828, VCV001120244, VCV002161909, VCV002429089, VCV003067137, VCV003391005. Available from: https://www.ncbi.nlm.nih.gov/clinvar/. Accessed 17 Oct 2026.
64. McClelland K, Li W, Rosenblum ND. Pallister-Hall syndrome, GLI3, and kidney malformation. Am J Med Genet C Semin Med Genet. 2022;190(3):264-278. doi:10.1002/ajmg.c.31999. PMID: 36165461.
65. Johnston JJ, Walker RL, Davis S, Facio F, Turner JT, Bick DP, et al. Zoom-in comparative genomic hybridisation arrays for the characterisation of variable breakpoint contiguous gene syndromes. J Med Genet. 2007;44(1):e59. doi:10.1136/jmg.2006.042473. PMID: 17098889. PMC2597909.
66. Kroisel PM, Petek E, Wagner K. Phenotype of five patients with Greig syndrome and microdeletion of 7p13. Am J Med Genet. 2001;102(3):243-9. doi:10.1002/ajmg.1443. PMID: 11484201.
67. Zung A, Petek E, Ben-Zeev B, Schwarzbraun T, Ben-Yehoshua SJ. MODY type 2 in Greig cephalopolysyndactyly syndrome (GCPS) as part of a contiguous gene deletion syndrome. Am J Med Genet A. 2011;155A(10):2469-72. doi:10.1002/ajmg.a.33829. PMID: 22043488.
68. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 82–83.
69. Elson E, Perveen R, Donnai D, Wall S, Black GC. De novo GLI3 mutation in acrocallosal syndrome: broadening the phenotypic spectrum of GLI3 defects and overlap with murine models. J Med Genet. 2002;39(11):804-6. doi:10.1136/jmg.39.11.804. PMID: 12414818. PMC1735022.
70. Speksnijder L, Cohen-Overbeek TE, Knapen MF, Lunshof SM, Hoogeboom AJ, van den Ouwenland AM, et al. A de novo GLI3 mutation in a patient with acrocallosal syndrome. Am J Med Genet A. 2013;161A(6):1394-400. doi:10.1002/ajmg.a.35874. PMID: 23633388.
71. Siafa L, Argilli E, Sherr EH, Myers KA. De Novo GLI3 Pathogenic Variants May Cause Hypotonia and a Range of Brain Malformations Without Skeletal Abnormalities. Pediatr Neurol. 2022;131:1-3. doi:10.1016/j.pediatrneurol.2022.03.005. PMID: 35436645. PMC10257559.
72. El Mouatani A, Van Winckel G, Zaafrane-Khachnaoui K, Whalen S, Achaiaa A, Kaltenbach S, et al. Homozygous GLI3 variants observed in three unrelated patients presenting with syndromic polydactyly. Am J Med Genet A. 2021;185(12):3831-3837. doi:10.1002/ajmg.a.62426. PMID: 34296525.
73. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 429–431.
74. Kariminejad A, Ghaderi-Sohi S, Keshavarz E, Hashemi SA, Parsimehr E, Szenker-Ravi E, et al. A GLI3 variant leading to polydactyly in heterozygotes and Pallister-Hall-like syndrome in a homozygote. Clin Genet. 2020;97(6):915-919. doi:10.1111/cge.13730. PMID: 32112393.
75. Green TE, Motelow JE, Bennett MF, Ye Z, Bennett CA, Griffin NG, et al. Sporadic hypothalamic hamartoma is a ciliopathy with somatic and bi-allelic contributions. Hum Mol Genet. 2022;31(14):2307-2316. doi:10.1093/hmg/ddab366. PMID: 35137044. PMC9307310.
76. Sami L, Chipaux M, Ferrand-Sorbets S, Doladilhe M, Bulteau C, Raffo E, et al. Genetic Insights Into Hypothalamic Hamartoma: Unraveling Somatic Variants. Neurol Genet. 2024;10(5):e200180. doi:10.1212/NXG.0000000000200180. PMID: 39246740. PMC11380500.
77. Green TE, Fujita A, Ghaderi N, Heinzen EL, Matsumoto N, Klein KM, et al. Brain mosaicism of hedgehog signalling and other cilia genes in hypothalamic hamartoma. Neurobiol Dis. 2023;185:106261. doi:10.1016/j.nbd.2023.106261. PMID: 37579995.
78. Green TE, Berkovic SF, Hildebrand MS. Somatic mosaicism in hypothalamic hamartoma. Curr Opin Genet Dev. 2026;99:102485. doi:10.1016/j.gde.2026.102485. PMID: 42155209.
79. Ng D, Johnston JJ, Turner JT, Boudreau EA, Wiggs EA, Theodore WH, et al. Gonadal mosaicism in severe Pallister-Hall syndrome. Am J Med Genet A. 2004;124A(3):296-302. doi:10.1002/ajmg.a.20338. PMID: 14708104.
80. Debeer P, Peeters H, Driess S, De Smet L, Freese K, Matthijs G, et al. Variable phenotype in Greig cephalopolysyndactyly syndrome: clinical and radiological findings in 4 independent families and 3 sporadic cases with identified GLI3 mutations. Am J Med Genet A. 2003;120A(1):49-58. doi:10.1002/ajmg.a.20018. PMID: 12794692.
81. Mukhtar NN, Demyati M, Benito A, Alzahrani AS. Case Report: Pallister-Hall syndrome diagnosed in adulthood during evaluation of recurrent hypoglycemia. Front Endocrinol (Lausanne). 2026;17:1925714. doi:10.3389/fendo.2026.1925714. PMID: 42856887. PMC13649492.
82. Bonilla-Navarrete S, Prieto LE, Carvajal LV, Olave-Rodríguez JA, Rodriguez-Rojas LX, Nastasi-Catanese JA. New Pathogenic Variant in the GLI3 Gene in the First Colombian Patient Associated With Pallister-Hall Syndrome: A Clinical Report. Mol Genet Genomic Med. 2025;13(10):e70146. doi:10.1002/mgg3.70146. PMID: 41063613. PMC12508622.
83. Saeed T, Khan B, Khan MH, Azeem Z, Fozia F, Ali S, et al. Expanding the Phenotypic and Genotypic Spectrum of Postaxial Polydactyly: A Study of Four Consanguineous Pakistani Families. Genet Test Mol Biomarkers. 2025 Dec 1. doi:10.1177/19450265251401006. PMID: 41338891.
84. Hong S, Han D, Qi M, Huang S, Ren M, Ding L, et al. Genetic landscape of pediatric seizures in Southeast China: identification of a novel GLI3 frameshift variant through whole-exome sequencing. Ann Med Surg (Lond). 2026;88(9):5612-9. doi:10.1097/MS9.0000000000005448. PMID: 42724907.
85. Tao S, Gu X, Wang X, Shen X, Zhao X. The molecular characterization of seven novel GLI family zinc finger 3 (GLI3) variants in Chinese families with limb malformations. Front Genet. 2025;16:1650790. doi:10.3389/fgene.2025.1650790. PMID: 41142227. PMC12545446.
86. Green TE, Schimmel M, Schubert S, Lemke JR, Bennett MF, Hildebrand MS, et al. Bi-allelic SMO variants in hypothalamic hamartoma: a recessive cause of Pallister-Hall syndrome. Eur J Hum Genet. 2022;30(3):384-388. doi:10.1038/s41431-021-01023-4. PMID: 35034092. PMC8904774.
87. Rubino S, Qian J, Pinheiro-Neto CD, Kenning TJ, Adamo MA. A familial syndrome of hypothalamic hamartomas, polydactyly, and SMO mutations: a clinical report of 2 cases. J Neurosurg Pediatr. 2019;23(1):98-103. doi:10.3171/2018.7.PEDS18292. PMID: 30497210.
88. Ondrey F, Griffith A, Van Waes C, Rudy S, Peters K, McCullagh L, et al. Asymptomatic laryngeal malformations are common in patients with Pallister-Hall syndrome. Am J Med Genet. 2000;94(1):64-7. doi:10.1002/1096-8628(20000904)94:1<64::aid-ajmg13>3.0.co;2-d. PMID: 10982485.
89. Azzam A, Lerner DM, Peters KF, Wiggs E, Rosenstein DL, Biesecker LG. Psychiatric and neuropsychological characterization of Pallister-Hall syndrome. Clin Genet. 2005;67(1):87-92. doi:10.1111/j.1399-0004.2005.00370.x. PMID: 15617553.
90. Boudreau EA, Liow K, Frattali CM, Wiggs E, Turner JT, Feuillan P, et al. Hypothalamic hamartomas and seizures: distinct natural history of isolated and Pallister-Hall syndrome cases. Epilepsia. 2005;46(1):42-7. doi:10.1111/j.0013-9580.2005.68303.x. PMID: 15660767.
91. Rahimian E, Tahsini MR, Azin N, Ghafouri-Fard S, Habibabadi JM, Fesharaki SSH, et al. An update in the MRI features, classification, and associated brain abnormalities in hypothalamic hamartomas. BJR Open. 2026;8(1):tzag015. doi:10.1093/bjro/tzag015. PMID: 42422568.
92. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 410–415.
93. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 168–171.
94. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 116–119.
95. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 390–393.
96. Jones KL, Jones MC, del Campo M. Hydrolethalus syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 254–255.
97. Jones KL, Jones MC, del Campo M. Oral-facial-digital syndrome type I. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 380–383.
98. Stoll C, De Saint Martin A, Donato L, Alembik K, Sauvage P, Messer J. Pallister-Hall syndrome with stenosis of the cricoid cartilage and microphallus without hypopituitarism. Genet Couns. 2001;12(3):231-5. PMID: 11693785.
99. Matsuyama M, Kobayashi H, Nakano S, Shimotakahara A, Hirai M. Subglottic stenosis associated with Pallister-Hall syndrome. Pediatr Int. 2024;66(1):e15816. doi:10.1111/ped.15816. PMID: 39417546.
100. Jamsheer A, Sowińska A, Trzeciak T, Jamsheer-Bratkowska M, Geppert A, Latos-Bieleńska A. Expanded mutational spectrum of the GLI3 gene substantiates genotype-phenotype correlations. J Appl Genet. 2012;53(4):415-22. doi:10.1007/s13353-012-0109-x. PMID: 22903559. PMC3477483.
101. Hurst JA, Jenkins D, Vasudevan PC, Kirchhoff M, Skovby F, Rieubland C, et al. Metopic and sagittal synostosis in Greig cephalopolysyndactyly syndrome: five cases with intragenic mutations or complete deletions of GLI3. Eur J Hum Genet. 2011;19(7):757-62. doi:10.1038/ejhg.2011.13. PMID: 21326280. PMC3128494.
102. Debeer P, Devriendt K, De Smet L, Deravel T, Gonzalez-Meneses A, Grzeschik KH, et al. The spectrum of hand and foot malformations in patients with Greig cephalopolysyndactyly. J Child Orthop. 2007;1(2):143-50. doi:10.1007/s11832-007-0022-8. PMID: 19308487. PMC2656707.
103. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 320–322.
104. Li H, Yang X, Liu L, Qin Y, Song Y, Yao Y. Genomic Landscape and Perinatal Outcomes of Fetal Polydactyly: A Retrospective Cohort Study Integrating CNV-seq and Trio-ES. Prenat Diagn. 2026;46(9):1364-73. doi:10.1002/pd.70216. PMID: 42390887.
105. Wang L, Li M, Dou G, Sun J, Zhou Z, Wang Y, et al. Dual molecular diagnosis of CEP290 and GLI3 mutations identified in an infant with leber congenital amaurosis and postaxial polydactyly, a Bardet-Biedl syndrome phenocopy. BMC Ophthalmol. 2026;26(1). doi:10.1186/s12886-026-04675-2. PMID: 41703510. PMC13020123.
106. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 416–419.
107. Niazi F, Goel K, Chen JS, Hadjinicolaou A, Keezer MR, Nguyen DK, et al. Outcome of Surgery for Hypothalamic Hamartoma-Related Epilepsy: A Systematic Review and Individual Participant Data Meta-Analysis. Neurology. 2024;103(12):e210060. doi:10.1212/WNL.0000000000210060. PMID: 39642321. PMC12334345.
108. Luisi C, Salimbene L, Pietrafusa N, Trivisano M, Marras CE, De Benedictis A, et al. Hypothalamic Hamartoma related epilepsy: A systematic review exploring clinical, neuropsychological, and psychiatric outcome after surgery. Epilepsy Behav. 2024;157:109846. doi:10.1016/j.yebeh.2024.109846. PMID: 38820683.
109. Garcia-Rodriguez R, Rodriguez-Rodriguez R, Garcia-Delgado R, Romero-Requejo A, Medina-Castellano M, Garcia Cruz L, et al. Prenatal diagnosis of Greig Cephalopolysyndactyly Syndrome. When to suspect it. J Matern Fetal Neonatal Med. 2022;35(11):2162-2165. doi:10.1080/14767058.2020.1774541. PMID: 32495660.
110. Raposo L, Fachada H, Santos Paulo A, Cerveira I, Castedo S, Pereira S. Prenatal diagnosis of Greig cephalopolysyndactyly syndrome: a case report. Prenat Diagn. 2015;35(2):203-5. doi:10.1002/pd.4506. PMID: 25297527.
111. Castro PT, Lopes J, Ribeiro G, Peixoto-Filho FM, Araujo Júnior E, Werner H. Prenatal diagnosis of Pallister-Hall syndrome: ultrasound, magnetic resonance imaging, and three-dimensional reconstructions of phenotypical findings. J Ultrasound. 2025;28(3):727-730. doi:10.1007/s40477-023-00782-8. PMID: 37067730. PMC12496359.
112. Dinu AM, Blumenfeld YJ, Smith CM. Shortened and Hypomineralized Bones, Renal Agenesis, and a Heart Defect: Prenatal Diagnosis of a GLI3 Variant. Prenat Diagn. 2025;45(13):1787-1790. doi:10.1002/pd.70007. PMID: 41178516.
113. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 58–61.
114. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 258–259.
115. Bornstein SR, Allolio B, Arlt W, Barthel A, Don-Wauchope A, Hammer GD, et al. Diagnosis and Treatment of Primary Adrenal Insufficiency: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2016;101(2):364-89. doi:10.1210/jc.2015-1710. PMID: 26760044. PMC4880116.
116. Jayakar P, Gaillard WD, Tripathi M, Libenson MH, Mathern GW, Cross JH, et al. Diagnostic test utilization in evaluation for resective epilepsy surgery in children. Epilepsia. 2014;55(4):507-18. doi:10.1111/epi.12544. PMID: 24512473.
117. Fleseriu M, Hashim IA, Karavitaki N, Melmed S, Murad MH, Salvatori R, et al. Hormonal Replacement in Hypopituitarism in Adults: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2016;101(11):3888-3921. doi:10.1210/jc.2016-2118. PMID: 27736313. Accepted manuscript consulted at: https://research.birmingham.ac.uk/en/publications/7d329c8c-9c14-4344-b113-49ce7f917271
118. Maeda-Usui A, Sato T, Nakano S, Kusakawa M, Kin T, Takahashi N, et al. Potential benefit of rapid genetic testing for Pallister-Hall syndrome. Clin Pediatr Endocrinol. 2023;32(2):119-122. doi:10.1297/cpe.2022-0065. PMID: 37020703. PMC10068620.
119. Elsherbini A, Perlmutter J, Mosa A. Genetic Syndromes Associated With Congenital Upper Limb Differences. J Hand Surg Am. 2026;51(5):553.e1-553.e10. doi:10.1016/j.jhsa.2026.01.008. PMID: 41689580.
120. Jones KL, Jones MC, del Campo M. Bardet-Biedl syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 826–829.
121. Bönnemann CG, Krishnamoorthy KS, Johnston JJ, Lee MM, Fowler DJ, Biesecker LG, et al. Clinical and molecular heterogeneity of syndromic hypothalamic hamartoma. Am J Med Genet A. 2023;191(9):2337-2343. doi:10.1002/ajmg.a.63306. PMID: 37435845. PMC10524239.
122. Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 200–201.
123. Jones KL, Jones MC, del Campo M. Acrocallosal syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 308–309.
124. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 242–245.
125. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 266–269.
126. Jones KL, Jones MC, del Campo M. Simpson-Golabi-Behmel syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 224–225.
127. Jones KL, Jones MC, del Campo M. 22q11.2 deletion syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 386–391.
128. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 212–215.
129. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 872–875.
130. Khan H, Bilal M, Bharadwaj T, Acharya A, Schrauwen I, Ahmad W, et al. HOXD12 a candidate gene for a novel form of synpolydactyly. Bone. 2026;207:117842. doi:10.1016/j.bone.2026.117842. PMID: 41786235.
131. Grimberg A, DiVall SA, Polychronakos C, Allen DB, Cohen LE, Quintos JB, et al. Guidelines for Growth Hormone and Insulin-Like Growth Factor-I Treatment in Children and Adolescents: Growth Hormone Deficiency, Idiopathic Short Stature, and Primary Insulin-Like Growth Factor-I Deficiency. Horm Res Paediatr. 2016;86(6):361-97. doi:10.1159/000452150. PMID: 27884013.
132. Grimberg A, Allen DB. Growth hormone treatment for growth hormone deficiency and idiopathic short stature: new guidelines shaped by the presence and absence of evidence. Curr Opin Pediatr. 2017;29(4):466-71. doi:10.1097/MOP.0000000000000505. PMID: 28525404. PMC5565215.
133. ALKINDI SPRINKLE (hydrocortisone) oral granules. Prescribing information. NDA 213876. Effective 2026-04-21. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=26c13a5f-7119-4c6a-bf10-fda4e07d7682
134. European Medicines Agency. Alkindi (hydrocortisone): EPAR — medicine overview. Marketing authorisation 2018-02-09. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/alkindi
135. European Medicines Agency. Efmody (hydrocortisone): EPAR. Marketing authorisation 2021-05-27. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/efmody
136. SOLU-CORTEF (hydrocortisone sodium succinate) for injection. Prescribing information. NDA 009866. Effective 2025-07-18. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=386942d3-c75f-25f1-e063-6394a90a5af2
137. AbbVie Inc. SYNTHROID (levothyroxine sodium) tablets, for oral use. Prescribing information. NDA 021402. Effective 2024-02-20. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=1e11ad30-1041-4520-10b0-8f9d30d30fcc
138. Pfizer. GENOTROPIN (somatropin) for injection. Prescribing information. BLA 020280. Effective 2026-08-04. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=ffebf88b-d257-4542-9808-74d9b7167765
139. Novo Nordisk. NORDITROPIN (somatropin) injection. Prescribing information. BLA 021148. Effective 2025-07-07. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=1058e17c-9261-459c-a3e6-fae38d196c14
140. European Medicines Agency. Omnitrope (somatropin): EPAR — product information. Authorisation 2006-04-12. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/omnitrope
141. Novo Nordisk. SOGROYA (somapacitan-beco) injection. Prescribing information. BLA 761156. Effective 2026-02-25. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=eadd03b4-adc3-4b88-8a24-9d205c5b85aa
142. European Medicines Agency. Sogroya (somapacitan): EPAR — product information. Authorisation 2021-03-31. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/sogroya
143. Pfizer. NGENLA (somatrogon-ghla) injection. Prescribing information. BLA 761184. Effective 2026-08-14. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=06dce529-b741-4a1b-8842-3ad77be6e308
144. European Medicines Agency. Ngenla (somatrogon): EPAR — product information. Authorisation 2022-02-14. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/ngenla
145. Ascendis Pharma. SKYTROFA (lonapegsomatropin-tcgd) for injection. Prescribing information. BLA 761177. Effective 2026-07-23. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=9229f871-bc61-4ddc-8122-d126d40e36ac
146. European Medicines Agency. Skytrofa (lonapegsomatropin): EPAR — product information. Authorisation 2022-01-11. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/skytrofa
147. Desmopressin acetate tablets. Prescribing information. ANDA 077414. Effective 2025-11-13. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=0d447b48-3f4b-430e-9e68-d8b9401c002e
148. Agencia Española de Medicamentos y Productos Sanitarios (CIMA). Minurin Flas 120 µg (reg. no. 68268), Minurin 0,1 mg/mL nasal spray solution (58566) and Minurin 4 µg/mL solution for injection (61413): SmPC (ficha técnica), section 4.2. Available from: https://cima.aemps.es/cima/publico/lista.html. Accessed 17 Oct 2026.
149. Desmopressin acetate nasal spray solution. Prescribing information. ANDA 074830. Effective 2025-06-19. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=8034989a-45f5-436b-8b1f-42108ee2a21a
150. Fresenius Kabi. Desmopressin acetate injection. Prescribing information. ANDA 216922. Effective 2023-09-06. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=0d96944d-736d-4b56-b726-d21eacd8bc85
151. Kübler K, Klingmüller D, Gembruch U, Merz WM. High-risk pregnancy management in women with hypopituitarism. J Perinatol. 2009;29(2):89-95. doi:10.1038/jp.2008.116. PMID: 19177043.
152. Wu X, Li Y, Ma H, Chen Q, Zhang J, Guo S, et al. Prevalence and effects on outcomes of intracranial tumors and structural alterations in children with central precocious puberty and early and fast puberty. Front Endocrinol (Lausanne). 2026;17:1880616. doi:10.3389/fendo.2026.1880616. PMID: 42568473.
153. Al Juraibah FN, Alharbi MM, Alharbi RD, Albaraki JS, AlDharman SS, Alghamdi HA, et al. Cranial MRI in Girls With Central Precocious Puberty: Clinical Characteristics and Predictors of Abnormal Findings. Saudi Med J. 2026;47(10):1657-64. doi:10.15537/1658-3175.8855. PMID: 42761971.
154. Kontbay Çetin T, Açıkgöz EG. Clinical and Etiological Evaluation of Male Patients with Central Precocious Puberty: Findings from a Single-Center Cohort. Turk Arch Pediatr. 2026;61(8):683-9. doi:10.65717/TurkArchPediatr.2026.25562. PMID: 42541464.
155. Wang X, Dou X, Zhang Q, Song D, Gong C. Reflections on GnRH analogues dosage in the context of escape phenomenon: a case report of hypothalamic hamartoma with central precocious puberty and literature review. J Pediatr Endocrinol Metab. 2026;39(10):1051-9. doi:10.1515/jpem-2025-0727. PMID: 42381421.
156. AbbVie Inc. LUPRON DEPOT-PED (leuprolide acetate for depot suspension), for intramuscular use. Prescribing information. NDA 020263. Revised 2025-11 [effective 2025-11-14]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=e99f47d2-da10-3127-ecb3-e5d942ae6e81
157. Tolmar Inc. FENSOLVI (leuprolide acetate) for injectable suspension, for subcutaneous use. Prescribing information. NDA 213150. Effective 2025-09-25. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=2c311700-edf4-4f2a-a468-9875489a6cc7
158. TRIPTODUR (triptorelin) for extended-release injectable suspension, for intramuscular use. Prescribing information. NDA 208956. Effective 2025-09-14. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=f41380e7-b830-432d-a5f5-a872932f107e
159. SUPPRELIN LA (histrelin acetate) subcutaneous implant. Prescribing information. NDA 022058. Effective 2025-09-25. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=d8fb000e-3cc9-4803-b71d-2cc597661977
160. Agencia Española de Medicamentos y Productos Sanitarios (CIMA). Decapeptyl Mensual 3,75 mg (reg. no. 58404), Trimestral 11,25 mg (61665) and Semestral 22,5 mg (72077): SmPC (ficha técnica), sections 4.1, 4.2 and 4.4. Available from: https://cima.aemps.es/cima/publico/lista.html. Accessed 17 Oct 2026.
161. Agencia Española de Medicamentos y Productos Sanitarios (CIMA). Gonapeptyl Depot 3,75 mg (reg. no. 64530): SmPC (ficha técnica), sections 4.1, 4.2 and 4.4. Available from: https://cima.aemps.es/cima/publico/lista.html. Accessed 17 Oct 2026.
162. Agencia Española de Medicamentos y Productos Sanitarios (CIMA). Lutrate Depot Mensual 3,75 mg (reg. no. 83377) and Trimestral 22,5 mg (83376): SmPC (ficha técnica), sections 4.1, 4.2 and 4.4. Available from: https://cima.aemps.es/cima/publico/lista.html. Accessed 17 Oct 2026.
163. Agencia Española de Medicamentos y Productos Sanitarios (CIMA). Ginecrin Depot 3,75 mg (reg. no. 59344; not marketed): SmPC (ficha técnica), sections 4.1, 4.2 and 4.4. Available from: https://cima.aemps.es/cima/publico/lista.html. Accessed 17 Oct 2026.
164. Agencia Española de Medicamentos y Productos Sanitarios (CIMA). Procrin Semestral 30 mg (reg. no. 74095): SmPC (ficha técnica), sections 4.2 and 4.4. Available from: https://cima.aemps.es/cima/publico/lista.html. Accessed 17 Oct 2026.
165. Carel JC, Eugster EA, Rogol A, Ghizzoni L, Palmert MR, Antoniazzi F, et al. Consensus statement on the use of gonadotropin-releasing hormone analogs in children. Pediatrics. 2009;123(4):e752-62. doi:10.1542/peds.2008-1783. PMID: 19332438.
166. Bangalore Krishna K, Fuqua JS, Rogol AD, Klein KO, Popovic J, Houk CP, et al. Use of Gonadotropin-Releasing Hormone Analogs in Children: Update by an International Consortium. Horm Res Paediatr. 2019;91(6):357-372. doi:10.1159/000501336. PMID: 31319416.
167. Gogou M, Barnes N, Carter S, Elliott C, D'Arco F, Gan HW, et al. Appetite and body weight changes after MR-guided laser interstitial thermal therapy of hypothalamic hamartomas in children with drug-resistant epilepsy. Seizure. 2026;137:7-16. doi:10.1016/j.seizure.2026.02.014. PMID: 41747505.
168. Wang X, Hu W, Zhang C, Zheng Z, Liu C, Sang L, et al. MRgLITT for hypothalamic hamartomas: surgical outcomes & divergent pathways for seizure prognosis. J Adv Res. 2026 Jul 13. doi:10.1016/j.jare.2026.07.016. PMID: 42442538.
169. Jehi L, Jette N, Kwon CS, Josephson CB, Burneo JG, Cendes F, et al. Timing of referral to evaluate for epilepsy surgery: Expert Consensus Recommendations from the Surgical Therapies Commission of the International League Against Epilepsy. Epilepsia. 2022;63(10):2491-506. doi:10.1111/epi.17350. PMID: 35842919.
170. Hussain N, Khalid MU, Szpindel A, Bouthillier A, Tahir MZ, Mirza FA. Surgical Treatment of Hypothalamic Hamartoma Causing Refractory Epilepsy: A Systematic Review. Can J Neurol Sci. 2026;53(5):718-723. doi:10.1017/cjn.2025.10385. PMID: 40776824.
171. Hajikarimloo B, Tos SM, Alvani MS, Kooshki A, Mohammadzadeh I, Zare AH, et al. Efficacy and safety of stereotactic radiosurgery for hypothalamic hamartomas: a systematic review and meta-analysis. Radiat Oncol. 2025;20(1):177. doi:10.1186/s13014-025-02754-0. PMID: 41299752.
172. Reinacher PC, Jacobs J, Shah MJ, Demerath T, Wagner K, San Antonio-Arce V, et al. Long-term outcomes of stereotactic radiofrequency ablation in hypothalamic hamartomas: A single-center experience. Epilepsia. 2026;67(1):96-108. doi:10.1111/epi.18660. PMID: 41030180.
173. Cossu G, Sufianov R, Messerer M, Ryvlin P, Daniel RT, Sufianov A. Intraventricular Endoscopic Surgery for Pediatric Hypothalamic Hamartomas: Surgical Series and Literature Meta-Analysis. World Neurosurg. 2025;204:124505. doi:10.1016/j.wneu.2025.124505. PMID: 41005656.
174. Cavalheiro S, Padua WL, Nôvo PC, de Souza Melo TC, Dastoli PA, Suzuki F, et al. Endoscopic ultrasonic aspirator-assisted resection for pediatric hypothalamic hamartomas: outcomes and technical nuances. Childs Nerv Syst. 2025;41(1):340. doi:10.1007/s00381-025-07014-w. PMID: 41177886.
175. De Benedictis A, Luisi C, Cappelletti S, Mercier M, Rossi-Espagnet MC, Luglietto D, et al. Endoscopic disconnection of hypothalamic hamartoma: an already outdated story? J Neurosurg Pediatr. 2026;37(6):593-604. doi:10.3171/2025.11.PEDS25377. PMID: 41962169.
176. Lallas M, Fajardo M, Resnick T, Duchowny M, Jayakar P, Altman N, et al. MR guided focused ultrasound ablation in epilepsy due to hypothalamic hamartoma: an initial experience [abstract 1229]. American Epilepsy Society Annual Meeting; 2025 Dec 7. Available from: https://aesnet.org/abstractslisting/mr-guided-focused-ultrasound-ablation-in-epilepsy-due-to-hypothalamic-hamartoma-an-initial-experience
177. InSightec. A Feasibility Safety Study of Benign Centrally-Located Intracranial Tumors in Pediatric and Young Adult Subjects. ClinicalTrials.gov identifier NCT03028246. Available from: https://clinicaltrials.gov/study/NCT03028246
178. Jacobs J, Hildebrand M. New insights to hypothalamic hamartoma syndrome. Epilepsy Behav. 2025;170:110468. doi:10.1016/j.yebeh.2025.110468. PMID: 40393150.
179. Kanjilal S, Bhaisora KS, Rai S, Kumar A, Maurya VP, Das KK, et al. Disconnection of Hypothalamic Hamartoma Through Trans-Sylvian Route in a Patient With Pallister-Hall Syndrome: 2-Dimensional Operative Video. Oper Neurosurg. 2026;30(5):796. doi:10.1227/ons.0000000000001764. PMID: 40910774.
180. National Human Genome Research Institute. Phenotype and Etiology of Pallister-Hall Syndrome. ClinicalTrials.gov identifier NCT00001404. Available from: https://clinicaltrials.gov/study/NCT00001404
181. Debiopharm International. LIBELULA: Debio 4326 (12-month triptorelin) in pediatric central precocious puberty. ClinicalTrials.gov identifier NCT06129539. Available from: https://clinicaltrials.gov/study/NCT06129539. Accessed 17 Oct 2026.
182. CASPPIAN: FP-001 42 mg (leuprolide mesylate depot) in central precocious puberty. ClinicalTrials.gov identifier NCT05493709. Available from: https://clinicaltrials.gov/study/NCT05493709. Accessed 17 Oct 2026.
183. Fondation Ophtalmologique Adolphe de Rothschild. Genetic and electrophysiologic study in focal drug-resistant epilepsies. ClinicalTrials.gov identifier NCT02890641. Available from: https://clinicaltrials.gov/study/NCT02890641. Accessed 17 Oct 2026.
184. Erasmus Medical Center. GROWing up with rare GENEtic syndromes. ClinicalTrials.gov identifier NCT04463316. Available from: https://clinicaltrials.gov/study/NCT04463316. Accessed 17 Oct 2026.
185. ClinicalTrials.gov. Records NCT05029622 (6-monthly triptorelin, Ipsen), NCT06926933 (leuprorelin 45 mg, HanAll), NCT05341115 (depot leuprorelin, Takeda), NCT07710430 (monthly leuprorelin, LG Chem) and NCT07864116 (basal LH threshold, Strasbourg). Available from: https://clinicaltrials.gov/. Accessed 17 Oct 2026 (API v2).
186. Damião SQ, Yuan J, Mittal A, Kearns C, Fung KFK. Pallister-Hall Syndrome. Radiographics. 2026;46(6):e250257. doi:10.1148/rg.250257. PMID: 42166346.
187. Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 291–293.