Pallister-Hall Syndrome

A number sign (#) is used with this entry because of evidence that Pallister-Hall syndrome (PHS) is caused by heterozygous mutation in the GLI3 gene (165240) on chromosome 7p14.

Description

Pallister-Hall syndrome is a pleiotropic autosomal dominant disorder comprising hypothalamic hamartoma, pituitary dysfunction, central polydactyly, and visceral malformations (Biesecker et al., 1996).

Clinical Features

Hall et al. (1980) reported 6 infants with a neonatally lethal malformation syndrome of hypothalamic hamartoblastoma, postaxial polydactyly, and imperforate anus. Some had laryngeal cleft, abnormal lung lobation, renal agenesis or dysplasia, short fourth metacarpals, nail dysplasia, multiple buccal frenula, hypoadrenalism, microphallus, congenital heart defect, and intrauterine growth retardation. All cases were sporadic and chromosomes were apparently normal. The parents were nonconsanguineous. No environmental exposure was common to all cases. The ages of the fathers were 21, 25, 25, 29, 43, and unknown. The anterior pituitary was not found in any patient. The hypothalamic tumor was apparent on the inferior surface of the cerebrum and extended from the optic chiasma to the interpeduncular fossa. The tumor replaced the hypothalamus and other nuclei that originate in the embryonic hypothalamic plate. It was composed mainly of cells resembling primitive, undifferentiated germinal cells.

Graham et al. (1983) described an infant with abnormal auricles, short nose with flattened bridge, microglossia, micrognathia, cleft palate, short limbs, dislocated hips, and 4-limb postaxial polydactyly. The infant died at 2 hours of age and autopsy showed hypothalamic hamartoblastoma. A sister of the mother died at 17 hours of age and showed 4-limb polydactyly, recessed mandible, and small tongue; autopsy was not done. See 241800.

Iafolla et al. (1989) reported that nail dysplasia accompanies the postaxial polydactyly.

Pallister et al. (1989) described 3 additional cases. One patient had imperforate anus and right hydronephrosis and hydroureter with absent left kidney; another had imperforate anus without the renal anomalies. A teratogen had been suspected because of the fact that cases were not recognized before February 1978. Pallister et al. (1989) suggested, however, that autopsy of cases of imperforate anus with associated CNS lesions should be reviewed.

Finnigan et al. (1991) described 2 unrelated patients with the usual features of Pallister-Hall syndrome, including diencephalic anomalies, but without hamartoblastomas.

Topf et al. (1993) reported affected father and son. The 9-year-old son had precocious puberty, imperforate anus, postaxial polydactyly, hypospadias, a hypothalamic mass, and a displaced pituitary gland; his 34-year-old father had polydactyly, a hypothalamic mass, and a flattened pituitary gland. In the son imperforate anus, postaxial polydactyly, and hypospadias, were surgically corrected early in life. His subsequent growth and development were normal. He had minor craniofacial dysmorphism. At the age of 8 years, he presented because of precocious puberty and was found by MRI scan to have a large hypothalamic mass and displaced pituitary. The father's polydactyly consisted of an accessory digit, apparently originating from the third metacarpal. His facial features resembled those of the son. He did not have a history of imperforate anus or hypospadias.

Penman Splitt et al. (1994) also reported an instance of male-to-male transmission: a 9-year-old boy had hamartoblastoma, midaxial and postaxial polydactyly with brachydactyly, hypospadias, imperforate anus, and precocious puberty. The 34-year-old father had central and postaxial polydactyly, macrocephaly, and a hypothalamic mass. Both were of normal intelligence.

Thomas et al. (1994) reported a definite sib recurrence of Pallister-Hall syndrome in a family without a cytogenetically visible chromosome abnormality. The father of the 2 affected sibs was born with nearly identical digital abnormalities and could represent either mild expression or mosaicism for a dominant gene. The first-born infant, a female, was noted at birth to have choanal atresia, bifid epiglottis, and cleft upper larynx with posterior web in the subglottic area. Her nose was short, with depressed nasal bridge and anteverted nares. There was bilateral hexadactyly with osseous 2/3 syndactyly of the right hand that was thought to represent an insertional type of polydactyly. Subsequently she developed signs of hypopituitarism, had recurrent problems with upper airway obstruction eventually necessitating tracheostomy, and required gastrostomy because of feeding difficulties. She died at 12 months of age after a respiratory infection. At necropsy, a tracheal diverticulum and abnormal lobation of the lungs were found. The pituitary gland was absent with a rudimentary stalk. A 2 x 2 cm soft nodular mass completely replaced the hypothalamus. The male sib was born at 30 weeks' gestation and required respiratory support after birth for upper airway obstruction. He had 4/5 syndactyly of the right hand with postaxial polydactyly of both hands and short fourth and fifth digits. He also had micropenis. Laryngoscopy and bronchoscopy showed a bifid uvula and epiglottis with cleft larynx. He subsequently was shown to have hypopituitarism and a cerebral MRI scan showed a 2 x 2 cm mass in the hypothalamic region with signal characteristics identical to those of normal brain tissue. He died suddenly at 9.5 months of age after a respiratory infection. The father was born with nearly complete 4/5 cutaneous syndactyly of the right hand and postaxial polydactyly of both hands. He had no other abnormalities. Both parents also had normal chromosomes.

Sama et al. (1994) reported an affected newborn female with a large suprasellar mass with a posterior cystic component, bilateral choanal atresia, renal hypoplasia, exomphalos, postaxial polydactyly, and underdevelopment of the 4th metacarpal. Her father was 38 years old.

Verloes et al. (1995) reported the cases of 2 unrelated, long surviving patients, aged 2 and 17 years, with Pallister-Hall syndrome. In addition to hypothalamic hamartoblastoma, both showed mild facial dysmorphism (downward slanted palpebral fissures, ptosis, microretrognathia), cleft epiglottis, and developmental delay. The younger child had stenosis of the pulmonary arteries, complex urogenital malformations, and anal atresia. The older patient had precocious puberty caused by the hamartoma, combined with complete growth hormone deficiency. Both patients showed bony anomalies of the limbs: variable proximal synostosis between the second to fourth metacarpals or intercalary polydactyly with generalized brachydactyly, severe brachytelephalangism, syndactyly, and nail hypoplasia.

The range of phenotypic variability in the Pallister-Hall syndrome may be demonstrated by the mother and son reported by Low et al. (1995). A 53-year-old woman and her 20-year-old son both presented with polysyndactyly but without other external malformations or mental retardation. MRI revealed, as an incidental finding, asymptomatic hypothalamic hamartomas in both patients. The sibs of both the mother and the son were unaffected.

Kang et al. (1997) described the clinical characteristics of a family with a mild form of PHS. Clinical, radiographic, and endoscopic evaluations showed that the disorder was fully penetrant with variable expressivity and low morbidity. The proband was a 21-month-old child who was noted to have ptosis and polydactyly at the time of birth. Cranial MRI showed a hypothalamic mass compatible with hamartoma. The family history showed that she had 21 relatives with polydactyly in an autosomal dominant pattern. The pedigree included 9 obligate heterozygotes, all of whom had some manifestation of the disorder. The ratio of males to females was 6 to 16. Several members of the family had central polydactyly (also referred to as mesoaxial or insertional polydactyly). This form typically includes partial osseous syndactyly with proximal fusion of the metacarpals and is characteristic of only a few syndromes.

Galasso et al. (2001) described a boy with Pallister-Hall syndrome with short stature and growth hormone neurosecretory dysfunction. The patient was found to have deficient spontaneous growth hormone secretion despite a normal response to pharmacologic stimulation. The patient responded positively to long-term growth hormone treatment. Galasso et al. (2001) concluded that growth hormone deficiency should be considered a cause of short stature in patients with Pallister-Hall syndrome, especially when their growth rates decrease.

Demurger et al. (2015) reported the molecular and clinical results from their study of a cohort of 76 probands with either a GLI3 mutation (49 with Greig cephalopolysyndactyly syndrome (GCPS; 175700) and 21 with PHS) or a large deletion encompassing the GLI3 gene (6 with GCPS). The diagnosis of PHS was confirmed in 4 individuals without polydactyly. Fetal cases with PHS showed severe craniofacial anomalies (agnathia, absence of oral orifice, cleft palate, or premaxillary agenesis) as well as reductional limb defects, suggesting the possible lethality of PHS. These severe cases revealed mutations in the middle third of the GLI3 gene between nucleotides 2941 and 3324. In situ hybridization confirmed early expression of GLI3 in pharyngeal arches, with later expression in the mandible and maxillary regions.

Inheritance

Autosomal dominant inheritance of Pallister-Hall syndrome is supported by the finding of male-to-male transmission (Kletter and Biesecker, 1992; Topf et al., 1993; Penman Splitt et al., 1994). Kletter and Biesecker (1992) stated that the Pallister-Hall syndrome is due to an autosomal dominant gene, that most cases are sporadic and the result of new mutation, and that the gene manifests variable expressivity.

Cytogenetics

Kuller et al. (1992) reported 3 cases. Two of them were affected brothers, one of whom had a chromosome abnormality: the father was a balanced carrier for the translocation t(3;7)(p25.3;q36); the offspring had an unbalanced karyotype with a der(7) chromosome ('7q+'). The parents, of Filipino extraction, were not known to be related. The infant, who died at age 5 weeks of necrotizing enterocolitis, had cleft lip and palate, agenesis of the nasal septum, hypopituitarism by endocrine evaluation, and fused thalami and a hypothalamic hamartomatous lesion by MRI. Postaxial polydactyly was not present and the anus was patent. The older brother died at the age of 3 years. A nonbanded karyotype performed in the Philippines was said to have been normal. Micropenis was present in both brothers. Kuller et al. (1992) suggested that the gene for this disorder may lie on either chromosome 3 or chromosome 7.

Mapping

After conducting a genomewide search with highly polymorphic markers in 4 families with PHS, including 32 affected and 36 unaffected persons, Kang et al. (1996) found linkage initially to D7S672. Physical mapping data from this region resulted in refinement of marker order. Genotyping with additional markers resulted in the maximum 2-point lod score between PHS and D7S691 of 8.04 at theta = 0.0. Multipoint analysis showed that the phenotype mapped between D7S521 and D7S678 with a peak lod score of 13.6 at D7S691. This region contains at least 2 candidate genes, GLI3 (165240) and inhibin beta-1 (147290). These candidate genes had previously been mapped to the 7p15-p13 region.

Using anonymous STRP markers in 2 'mildly affected' PHS families, Kang et al. (1997) established linkage to 7p13 by 2-point analysis with D7S691, resulting in a lod score of 7.0 at theta = 0.0, near the GLI3 locus.

Diagnosis

Iafolla et al. (1989) pointed out that magnetic resonance imaging is the most valuable diagnostic tool; CT scan has been reported to miss the tumor.

An international workshop on Pallister-Hall syndrome (Biesecker et al., 1996) developed minimal diagnostic criteria for this entity. The index case in a family must have both hypothalamic hamartoma and central polydactyly to meet the diagnostic criteria. First-degree relatives of the index case must have either hypothalamic hamartoma or polydactyly (central or postaxial) and show inheritance in an autosomal dominant pattern or in a manner consistent with gonadal mosaicism. Recommendations for clinical evaluation of suspected cases were presented. Biesecker et al. (1996) concluded that hypothalamic hamartoma is not specific to PHS.

Prenatal Diagnosis

Sills et al. (1993) reported Pallister-Hall syndrome in a male infant and his female sib fetus. Sills et al. (1994) reported that the parents elected to terminate their third pregnancy because prenatal ultrasonographic findings suggested PHS. At 12 weeks, ultrasound demonstrated an abnormal-appearing fetus with large head and a large midline anterior fluid-filled structure, possibly representing a dilated, malformed ventricle. Repeat examination at 13 weeks showed evidence of holoprosencephaly, enlarged cisterna magna, short umbilical cord, polydactyly, and possible syndactyly. The limbs appeared short. At termination at 15 weeks, chromosome studies of chorionic villus cells showed a normal 46,XX karyotype.

Differential Diagnosis

Verloes et al. (1992) commented that 'Some years ago, syndromologists and clinical geneticists were pleasantly divided into splitters...and lumpers...' With the increasing volume of reported congenital anomalies, they suggested that 2 new categories may be delineated: the 'stretchers,' who are mainly preoccupied with extending the limits of a given phenotype by adding milder or unusually severe variants, and the 'cut-and-pasters,' who displace the same atypical or borderline cases from one syndrome to another. Verloes et al. (1992) reviewed the differential diagnosis of Pallister-Hall syndrome and of syndromal hypothalamic hamartoblastoma in general. The conditions that they considered as related included Smith-Lemli-Opitz syndrome (270400), the pseudotrisomy 13 syndrome or holoprosencephaly-polydactyly syndrome (264480), orofaciodigital syndrome type VI or Varadi-Papp syndrome (277170), and the hydrolethalus syndrome (236680). Because of the insolvable overlap, Verloes et al. (1992) suggested the creation of a phenotypic classification called the multiplex syndrome, defined as a clinical frame encompassing an unknown number of genetic and/or nongenetic multiple congenital anomaly (MCA) syndromes, for which differential diagnosis cannot be performed unequivocally. They suggested calling this the cerebroacrovisceral early lethality (CAVE) multiplex syndrome.

Verloes et al. (1995) described a 24-week-old fetus with agenesis of the corpus callosum, arhinencephaly, hypothalamic hamartoblastoma, absence of the right thumb, hypoplastic left thumb, hypoplastic lungs, intestinal malrotation, microgastria, asplenia, 'inverted' horseshoe kidney, blind vagina, and absence of internal genitalia. Underdevelopment of preaxial structures had not previously been reported in the Pallister-Hall syndrome, in which postaxial polydactyly is a cardinal feature. Although some features suggested the 'microgastria-limb reduction' complex (156810), patients with that complex are usually not so severely affected and have never had associated hypothalamic hamartoblastoma.

Donnai et al. (1987) had suggested that the Pallister-Hall syndrome and severe Smith-Lemli-Opitz syndrome (270400) are the same disorder. Using multivariate analysis and numerical taxonomy, Verloes et al. (1995) concluded that on review, 'most overlapping cases (and, in fact, most cases reported as Pallister-Hall, including some from the original report)' could be unambiguously classified as Smith-Lemli-Opitz syndrome, orofaciodigital syndrome type VI (277170), or holoprosencephaly-polydactyly syndrome (264480). Together with the absence of anomalies of cholesterol metabolism, a combination of oral frenula, laryngeal malformations, digestive abnormalities, intercalary polysyndactyly, generalized brachytelephalangism, and nail hypoplasia should allow the delineation of Pallister-Hall syndrome, even when a CNS tumor is absent (Verloes, 1995). The radiologic abnormalities in the hand are helpful in differentiating Pallister-Hall syndrome from other syndromes in which hypothalamic hamartoblastoma is observed. This may be of major importance for genetic counseling because Pallister-Hall syndrome is an autosomal dominant disorder, whereas most most of the other disorders with the CAVE phenotype are recessively inherited.

Because of a possible relationship of PHS to Smith-Lemli-Opitz syndrome, Biesecker et al. (1996) analyzed levels of cholesterol and intermediate metabolites of the later stages of cholesterol biosynthesis and found no evidence of a generalized disorder of cholesterol biosynthesis in patients with familial PHS. On genetic and biochemical grounds, they concluded that PHS and Smith-Lemli-Opitz syndrome are not allelic variants of a single locus.

Unsinn et al. (1995) described a patient who had hydrocolpos and postaxial polydactyly as well as hypothalamic hamartoblastoma, raising the possibility that the Pallister-Hall syndrome and the McKusick-Kaufman syndrome (MKKS; 236700) are one entity. Lurie (1995) questioned whether the Kaufmann-McKusick syndrome and the Pallister-Hall syndrome can be considered one entity. Lurie and Wulfsberg (1994) found in a survey of 43 reported familial cases of MKKS neither preaxial nor central forms of polydactyly. Lurie (1995) stated that he also did not remember vaginourethral fistula as occurring in these patients.

Kang et al. (1997) noted that large deletions or translocations resulting in haploinsufficiency of the GLI3 gene have been associated with Greig cephalopolysyndactyly syndrome (GCPS; 175700), which maps to the same region of chromosome 7p, although no mutations in GLI3 were identified in GCPS patients with normal karyotypes. Both PHS and GCPS have polysyndactyly and abnormal craniofacial features and are inherited in an autosomal dominant pattern, but are clinically distinct. The polydactyly of GCPS is commonly preaxial and that of PHS is typically central or postaxial. No reported cases of GCPS had hypothalamic hamartoma and PHS does not cause hypertelorism or broadening of the nasal root or forehead.

Some patients with PHS have a bifid epiglottis, a rare malformation. On laryngoscopy in 26 subjects with PHS, Ondrey et al. (2000) found that 15 had a bifid or cleft epiglottis (58%); of 14 subjects with GCPS, which is also caused by mutation in the GLI3 gene, no instance of cleft epiglottis was found. Malformed epiglottis was asymptomatic in all of the prospectively evaluated subjects. One additional PHS subject was found to have bifid epiglottis and a posterior laryngeal cleft on autopsy. Ondrey et al. (2000) concluded that bifid epiglottis is common in PHS, whereas posterior laryngeal clefts are uncommon and occur only in severely affected patients. They suggested that the diagnosis of a bifid epiglottis should prompt a thorough search for other sometimes asymptomatic anomalies of PHS to provide better medical care and recurrence risk assessment for affected individuals and families.

Molecular Genetics

Because of the colocalization of the loci for PHS and GCPS, Kang et al. (1997) investigated GLI3 as a candidate gene for PHS. They reported 2 PHS families with frameshift mutations in GLI3 (165240.0002; 165240.0003) located 3-prime of the zinc finger-encoding domains, including 1 family with a de novo mutation.

Killoran et al. (2000) reported a family in which the proposita was born with central polydactyly of the right hand, broad thumbs, and short fingers with hypoplastic nails. She also had a small right kidney, imperforate anus, and a hypoplastic iliac bone. An older sib with renal agenesis and polydactyly had died. The mother had polydactyly and imperforate anus. A maternal aunt had polydactyly, as did the maternal grandfather. Molecular analysis of the proposita identified a mutation in the GLI3 gene (165240.0017). The authors noted that subsequent clinical data showed that the patient had a hypothalamic hamartoma. This finding and the molecular analysis led to the diagnosis of PHS. Killoran et al. (2000) stated that they had originally diagnosed their patient with PIV (polydactyly, imperforate anus, and vertebral anomalies) syndrome as defined by Say and Gerald (1968). They doubted that the PIV designation was still useful and suggested that it be viewed as a historic antecedent to the VACTER/VACTERL association (192350).

Stoll et al. (2001) described a patient considered to have Pallister-Hall syndrome in whom they could not identify a mutation in the GLI3 gene. The patient had microphallus and no growth hormone deficiency. He had presented with postaxial polydactyly of the hands, dysplastic nails, imperforate anus, small penis, scrotum bifidum with very thin urethra, bifid epiglottis, and a bilateral simian crease. There was vesicoureteral reflux, insertional hexadactyly of the left hand, and 2 Y-shaped metacarpals with 6 fingers in the right hand. Brain MRI revealed a large sellar and suprasellar mass. Bronchoscopy showed anterior synechia of the vocal cords with cricoid stenosis. A tracheostomy was performed. Mental development was normal. In an erratum, the authors stated that they had identified a missense mutation in exon 13 of the GLI1 gene in this patient.

Genotype/Phenotype Correlations

Johnston et al. (2005) hypothesized that GLI3 mutations that predict a truncated functional repressor protein cause Pallister-Hall syndrome, whereas haploinsufficiency of GLI3 cause Greig cephalopolysyndactyly syndrome (GCPS; 175700). To test this hypothesis, they screened 46 patients with PHS and 89 patients with GCPS for GLI3 mutations. They detected 47 pathologic mutations (among 60 probands), and when these mutations were combined with previously published mutations, 2 genotype/phenotype correlations were evident. GCPS was caused by many types of alterations, including translocations, large deletions, exonic deletions and duplications, small in-frame deletions, and missense, frameshift/nonsense, and splicing mutations. In contrast, PHS was caused only by frameshift/nonsense and splicing mutations. Among the frameshift/nonsense mutations, Johnston et al. (2005) found a clear genotype/phenotype correlation. Mutations in the first third of the gene (from open reading frame nucleotides 1-1997) caused GCPS, and mutations in the second third of the gene (from nucleotides 1998-3481) caused primarily PHS. Surprisingly, there were 12 mutations in patients with GCPS in the 3-prime third of the gene (after open reading frame nucleotide 3481), and no patients with PHS had mutations in this region. These results demonstrated a robust genotype/phenotype correlation for GLI3 mutations and strongly supported the hypothesis that these 2 allelic disorders have distinct modes of pathogenesis.

Narumi et al. (2010) reported 2 Japanese patients, a 3-year-old girl and a 10-year-old boy, with PHS and genital anomalies, both of whom had heterozygous nonsense mutations in the GLI3 gene. The authors reviewed the 12 previously reported patients with PHS and genital anomalies, all of whom had nonsense or frameshift mutations in exons 13, 14, or 15 of GLI3; however, no hotspot for GLI3 mutations was found. In affected males, hypospadias was observed in 3 patients, micropenis in 2, and bifid or hypoplastic scrotum in 2. Three also had an anorectal abnormality (imperforate anus) and 1 had a urologic abnormality (small kidney). In affected females, all had hydrometrocolpos and/or vaginal atresia; 2 also had an anorectal abnormality (imperforate anus and rectoperineal fistula), and 4 had various urologic abnormalities, including vesicoureteral reflux in 2. Some of these patients had affected family members with anorectal anomalies and/or rectovaginal fistula, but no genital abnormalities. Narumi et al. (2010) concluded that genital features in PHS patients might present a wide range of severity even among individuals with the same nucleotide change. Noting that none of the female patients and only 2 of the male patients had panhypopituitarism, the authors suggested that the urogenital and anorectal abnormalities might be related to dysregulation of SHH (600725) signaling rather than hormonal aberrations.

Demurger et al. (2015) reported the molecular and clinical results from a study of 76 probands from 55 families who had either a mutation in GLI3 (49 with GCPS and 21 with PHS) or a large deletion encompassing GLI3 (6 with GCPS). Most of the mutations they identified were novel and supported previously reported genotype/phenotype correlations. Truncating mutations in the middle third of the gene generally resulted in PHS, whereas exonic deletions and missense and truncating mutations elsewhere in the gene caused GCPS.