Thanatophoric Dysplasia, Type I

A number sign (#) is used with this entry because of evidence that thanatophoric dysplasia type I (TD1) is caused by heterozygous mutation in the gene encoding the fibroblast growth factor receptor-3 (FGFR3; 134934) on chromosome 4p16.

Thanatophoric dysplasia type II (TD2; 187601), achondroplasia (ACH; 100800), and SADDAN (616482) are allelic disorders.

Description

Thanatophoric dysplasia is a severe short-limb dwarfism syndrome that is usually lethal in the perinatal period. Norman et al. (1992) classified cases of TD into subtypes based on the presence of curved as opposed to straight femurs; patients with straight, relatively long femurs always had associated severe cloverleaf skull and were designated TD type II (TD2), while TD cases with curved, short femurs with or without cloverleaf skull were designated TD type I (TD1) (Langer et al., 1987).

Clinical Features

Maroteaux et al. (1967) referred to patients with micromelic dwarfism who died in the first hours of life as having 'thanatophoric dwarfism.' The ribs and bones of the extremities were very short and the vertebral bodies were greatly reduced in height with wide intervertebral spaces; caudad narrowing of the spinal canal was not present. Radiologically, the vertebral bodies were H-shaped in frontal projection, and the femurs were shaped like telephone receivers. Maroteaux et al. (1967) found cases in the literature that matched this description, the earliest reported by Maygrier (1898).

Giedion (1968) described a Swiss patient with TD1 who had radioulnar synostosis and survived for 96 hours.

Stensvold et al. (1986) reported survival for 169 days. The child had increasing hydrocephalus and the femurs were shaped like telephone receivers. Tonoki (1987) described a patient who survived for 212 days.

MacDonald et al. (1989) reported unusually long survival; an unrelated boy and girl were still alive at the ages of 4.75 and 3.7 years, respectively. Both seemed to have disease of the usual severity. Surprisingly, the girl survived unsupported until age 2 months.

Knisely (1989) pointed out that megalencephaly and highly characteristic temporal lobe malformations are invariably present in thanatophoric dysplasia and that other abnormalities in central nervous system topography are frequently apparent by microscopy (Wongmongkolrit et al., 1983; Ho et al., 1984).

Baker et al. (1997) reported a patient with TD1 who survived beyond age 9 years. This patient also had acanthosis nigricans. The authors referred to another TD long-term survivor with acanthosis nigricans. This skin disorder also occurs in Crouzon syndrome (123500) when caused by a FGFR3 mutation (134934.0011). Genetic analysis in the patient of Baker et al. (1997) identified a common FGFR3 mutation (R248C; 134934.0005).

Pannier et al. (2009) reported a fetus with lethal TD1 ascertained at 24 weeks' gestation when the fetus was noted to have severe dwarfism. The pregnancy was terminated. Radiographic studies showed severe rhizomelic shortness of the long bones and mild bowing of the femora, radii, and ulnae. The spine showed severe platyspondyly with H-shaped vertebrae and narrowing of the interpediculate distance. The thorax was small with short ribs, and the iliac bones were short and wide. Macrocrania and frontal bossing were observed; there was no evidence of a cloverleaf skull. Postmortem examination showed cerebral cortical malformations with temporal lobe polymicrogyria and severe disorganization of growth plates in the long bones. Genetic analysis identified heterozygosity for 2 de novo missense mutations in the FGFR3 gene on the same allele (N540K and Q485R; 134934.0034). The authors noted that the N540K mutation in isolation (134934.0010) usually results in the less severe phenotype of hypochondroplasia (HCH; 146000).

Reviews

Sillence et al. (1978) provided a review of neonatal dwarfism, including thanatophoric dysplasia.

Inheritance

Maroteaux et al. (1967) concluded that a dominant mutation was the most likely basis for this disorder, but that recessive inheritance could not be excluded.

Pena and Goodman (1973) reviewed reported cases and concluded that polygenic inheritance was most likely. They suggested an empiric recurrence risk in sibs of 2%. McKusick (1973) suggested genetic heterogeneity, with some recessive and many dominant new mutation cases.

Bouvet et al. (1974) found no increase in parental age and no increase in parental consanguinity among patients with thanatophoric dwarfism. They pictured concordantly affected twins.

Serville et al. (1984) reported identical twins who were both affected, bringing to 4 the number of such cases. This was in distinct contrast to the rarity of affected nontwin sibs; only Bouvet et al. (1974) and Partington et al. (1971) had described such cases, and thanatophoric dysplasia was associated with cloverleaf skull in the latter cases, consistent with TD2. Corsello et al. (1992) observed thanatophoric dysplasia in monozygotic male twins, only 1 of whom had cloverleaf skull.

In a population study in the West of Scotland for the period 1970 to 1983, Connor et al. (1985) found that TD had an incidence of 1 in 42,221 births, which was consistent with a new dominant mutation rate of 11.8 +/- 4.1 x 10(-6) mutations per gene per generation.

In a collaborative study in Spain, Martinez-Frias et al. (1988) identified 13 cases among 517,970 births, an incidence of 2.7 per 100,000 births. All cases were sporadic, and there was no evidence of parental consanguinity. Parental age was increased; the average age of the fathers was 34.8, as compared to 35.3 in an equal number of sporadic achondroplasia cases, and 30.0 in 10,624 control births. The findings were interpreted as supporting autosomal dominant inheritance with an overall mutation rate of 1.34 per 100,000 gametes, a value close to that observed in achondroplasia.

Young et al. (1989) described concordantly affected female monozygotic twins. Monozygosity was established by DNA minisatellite fingerprinting.

Diagnosis

Prenatal Diagnosis

In utero diagnosis was demonstrated by Keats et al. (1970).

Although prenatal diagnosis of TD had been accomplished by ultrasonography in the second trimester (Schild et al., 1996), it was not always possible to distinguish between TD and other osteochondrodysplasias in utero.

Using restriction enzyme analysis, Sawai et al. (1999) identified a mutation in the FGFR3 gene in a fetus at 27 weeks' gestation.

Cytogenetics

Hersh et al. (1995) suggested that the gene mutant in TD might be on chromosome 1 or chromosome 10 because of the observation of a de novo 1;10 balanced translocation in an infant with this disorder. Although the possibility exists of genetic heterogeneity with one or more forms of TD due to mutations in a gene other than FGFR3, it is also possible that the balanced translocation was merely a coincidence in this case. The TD was classified as type I and there was no cloverleaf skull.

Molecular Genetics

Reardon et al. (1994) noted that fibroblast growth factor receptor-3 (FGFR3; 134934), which is mutant in achondroplasia, is structurally very similar to FGFR2. Their observation that FGFR2 mutations cause craniosynostosis suggested to them that the lethal skeletal disorder TD2 with cloverleaf skull (187601) may be a good candidate for further mutation searches in FGFR3. Furthermore, because of the phenotypic similarities between TD and homozygous achondroplasia, Arthur Beaudet independently suggested to Wasmuth (1995) that FGFR3 be studied in cases of TD. Indeed, Tavormina et al. (1995) found that 23 of 39 TD1 patients harbored amino acid substitutions in the extracellular domain of FGFR3. Of these, 22 patients were found to be heterozygous for the R248C substitution (134934.0005). One patient had a S371C substitution (134934.0006). Phenotypic heterogeneity was observed in TD1, in that 6 of 11 TD1 patients with cloverleaf skull and 16 of 28 patients without cloverleaf skull had the R248C mutation. Tavormina et al. (1995) found that all 16 patients with TD2 had a heterozygous K650E substitution (134934.0004). All 16 patients with TD2 had severe cloverleaf skull deformity with straight femurs. In a subsequent paper, Tavormina et al. (1995) identified a S249C mutation (134934.0013) in 4 cases of TD1. The authors proposed that the severe lethal phenotype in TD1 was more directly related to the introduction of a new cysteine residue than to the specific site of the amino acid substitution. They speculated that an unpaired cysteine residue in the cytoplasmic region of the protein may result in formation of intermolecular disulfide hybrids between two FGFR3 monomers, resulting in a constitutively active mutant receptor homodimer complex.

Rousseau et al. (1996) performed FGFR3 mutation analysis in 26 cases of TD1. Three missense mutations (Y373C, 134934.0016; R248C, 134934.0005; S249C, 134934.0013) accounted for 73% of the cases. Two stop codon mutations (X807R, 134934.0008; X807C, 134934.0009) and 1 rare G370C mutation were also found. Rousseau et al. (1996) noted that all reported missense mutations created cysteine residues and were located in the extracellular domain of the receptor. The findings provided support for the hypothesis that the newly created cysteine residues may allow disulfide bonds to form between the extracellular domains of mutant monomers, thus inducing constitutive activation of the homodimer receptor complex.

Brodie et al. (1999) examined 22 cases of thanatophoric dysplasia variants for the presence of missense mutations in FGFR3. All 17 cases of the San Diego variant (originally called the San Diego form of lethal short-limbed platyspondylic dysplasia) were heterozygous for the same FGFR3 mutations found in TD1; the R248C mutation was present in 7 of the 17 cases. Large inclusion bodies were found in all 14 cases of the San Diego type in which they were sought. Similar inclusion bodies were present in 2 of 72 thanatophoric dysplasia type I cases, but not in 39 controls. The material retained within the rough endoplasmic reticulum stained only with antibody to the FGFR3 protein. No mutations were identified in patients with the Torrance and Luton types of TD (151210). Brodie et al. (1999) suggested that the radiographic and cellular differences between thanatophoric dysplasia and the San Diego variant may be a consequence of other genetic factors, perhaps in the processing of mutant FGFR3 molecules within the rough endoplasmic reticulum. Hall (2002) noted that the San Diego form of platyspondylic lethal skeletal dysplasia (PLSD) described by Horton et al. (1979) had been classified as the same as thanatophoric dysplasia type I.

Genotype/Phenotype Correlations

Wilcox et al. (1998) examined the clinical, radiographic, and histologic findings in 91 cases with FGFR3 mutations from the International Skeletal Dysplasia Registry. The most common mutation was R248C (134934.0005), occurring in 45 (50%) cases, and the second most common mutation was Y373C (134934.0016), occurring in 18 cases (20%). All of these patients had TD1 characterized by curved femora and infrequent cloverleaf skull. All 17 (19%) patients with the K650E mutation (134934.0004) had TD2, characterized by straight femora with craniosynostosis and frequent cloverleaf skull. TD1 patients with the Y373C mutation tended to have more severe radiographic manifestations than TD1 patients with the R248C mutation, but there was phenotypic overlap between them. Histopathologically, all cases shared similar abnormalities, but those with the K650E mutation had better preservation of the growth plate.

Pathogenesis

Horton et al. (1988) delineated the process of abnormal ossification in 15 TD fetuses and infants from whom growth plate cartilage was obtained.

Delezoide et al. (1997) studied 18 thanatophoric dysplasia fetuses for FGFR3 expression in cartilage sections by in situ hybridization and immunohistochemistry. Specific antibodies revealed high amounts of FGFR3 in cartilage of TD fetuses with no increased level of the corresponding mRNA. The specific signal was mainly detected in the nucleus of proliferative and hypertrophic chondrocytes. Based on this observation and the abnormal expression of collagen type X (120110) in hypertrophic thanatophoric dysplasia chondrocytes, Delezoide et al. (1997) suggested that constitutive activation of the receptor through formation of a stable dimer increases its stability and promotes its translocation into the nucleus, where it might interfere with terminal chondrocyte differentiation.

Population Genetics

Camera and Mastroiacovo (1982) identified 13 cases of thanatophoric dysplasia among 217,061 Italian births. All were sporadic. In the same series, there were 8 cases of achondroplasia (100800) and 1 case each of camptomelic dysplasia (114290), Ellis-van Creveld syndrome (225500), Larsen syndrome (150250), and Langer mesomelic dysplasia (249700). Thanatophoric dysplasia was the most frequent skeletal dysplasia.

Connor et al. (1985) identified 43 cases of lethal neonatal short-limb chondrodysplasias in the West of Scotland for the period 1970 to 1983, representing a minimum incidence of 1 in 8,900. TD had an incidence of 1 in 42,221 births. The differential diagnosis included a number of well-delineated skeletal dysplasias: asphyxiating thoracic dysplasia and short-rib polydactyly (see 208500), achondrogenesis type II (200700), metatropic dysplasia (156530), OI congenita (166210), campomelic dysplasia (114290), rhizomelic chondrodysplasia punctata (215100), hypophosphatasia (241500), SED congenita (183900), one case of Warfarin embryopathy, and one apparently 'new' condition with presumed autosomal recessive inheritance (see 273680).

Orioli et al. (1986) estimated the frequency of TD to be approximately 1 in 20,000 births, making it the most common neonatal lethal skeletal dysplasia. In a collaborative study in Spain, Martinez-Frias et al. (1988) identified 13 cases among 517,970 births, an incidence of 2.7 per 100,000 births.

Between 1970 and 1983 in Denmark, Andersen (1989) found 2 cases of thanatophoric dysplasia among 77,977 births, including stillbirths. In addition, there was 1 case of thanatophoric dysplasia with cloverleaf skull.

Using data from 7 population-based birth defects monitoring programs in the United States, Waller et al. (2008) estimated the prevalence of achondroplasia and thanatophoric dysplasia and presented data on the association between older paternal age and these conditions. The prevalence of thanatophoric dysplasia ranged from 0.21 to 0.30 per 10,000 live births (1/33,330-1/47,620 live births). The prevalence of achondroplasia ranged from 0.36 to 0.60 per 10,000 live births (1/27,780-1/16,670 live births). These data suggest that thanatophoric dysplasia is one-third to one-half as frequent as achondroplasia. The differences on the prevalence of these conditions across monitoring programs were consistent with random fluctuation. In Texas, fathers that were 25-29, 30-34, 35-39, and over 40 years of age had significantly increased rates of de novo achondroplasia and thanatophoric dysplasia among their offspring compared with younger fathers.

History

Sabry (1974) reported affected triplets whose parents were first cousins. In reviewing the radiographs, however, Rimoin (1975) concluded that these sibs had achondrogenesis of the Parenti-Fraccaro type (200600). The affected sibs of normal parents, reported by Harris and Patton (1971), were subsequently concluded to have achondrogenesis (Harris et al., 1972). Chemke et al. (1971) and Graff et al. (1972) described thanatophoric dwarfism in 2 male offspring of first-cousin Moroccan Jewish parents. In the second-born affected sib the diagnosis was made antenatally by x-ray. However, after review of the radiographs of one, Rimoin (1975) concluded that this was not thanatophoric dwarfism. (Knowles et al. (1986) and Borochowitz et al. (1986) suggested that the disorder reported by Chemke et al. (1971) and Graff et al. (1972) was the same as the 'new' autosomal recessive dysplasia they described under the designation of Schneckenbecken dysplasia; see 269250.)