Achondroplasia

A number sign (#) is used with this entry because achondroplasia (ACH) is caused by heterozygous mutation in the fibroblast growth factor receptor-3 gene (FGFR3; 134934) on chromosome 4p16.3.

Description

Achondroplasia is the most frequent form of short-limb dwarfism. Affected individuals exhibit short stature caused by rhizomelic shortening of the limbs, characteristic facies with frontal bossing and midface hypoplasia, exaggerated lumbar lordosis, limitation of elbow extension, genu varum, and trident hand (summary by Bellus et al., 1995).

Clinical Features

Whereas many conditions that cause short stature have inappropriately been called achondroplasia in the past, the phenotype of this osteochondrodysplasia is so distinctive and so easily identified clinically and radiologically at birth that confusion should not occur. It is characterized by a long, narrow trunk, short extremities, particularly in the proximal (rhizomelic) segments, a large head with frontal bossing, hypoplasia of the midface and a trident configuration of the hands. Hyperextensibility of most joints, especially the knees, is common, but extension and rotation are limited at the elbow. A thoracolumbar gibbus is typically present at birth, but usually gives way to exaggerated lumbar lordosis when the child begins to ambulate. Mild to moderate hypotonia is common, and motor milestones are usually delayed. Intelligence is normal unless hydrocephalus or other central nervous system complications arise. In 13 achondroplastic infants, Hecht et al. (1991) found that cognitive development was average and did not correlate with motor development which typically was delayed. It was noteworthy that reduced mental capacity correlated with evidence of respiratory dysfunction detected by polysomnography.

In children, caudad narrowing of the interpediculate distance, rather than the normal caudad widening, and a notchlike sacroiliac groove are typical radiologic features. Also in children, epiphyseal ossification centers show a circumflex or chevron seat on the metaphysis. Limb shortening is especially striking in the proximal segments, e.g., the humerus; hence the description rhizomelic ('root limb'). The radiologic features of true achondroplasia and much concerning the natural history of the condition were presented by Langer et al. (1967) on the basis of a study of 101 cases and by Hall (1988).

True megalencephaly occurs in achondroplasia and has been speculated to indicate effects of the gene other than those on the skeleton alone (Dennis et al., 1961). Disproportion between the base of the skull and the brain results in internal hydrocephalus in some cases. The hydrocephalus may be caused by increased intracranial venous pressure due to stenosis of the sigmoid sinus at the level of the narrowed jugular foramina (Pierre-Kahn et al., 1980). Hall et al. (1982) pointed out that the large head of the achondroplastic fetus creates an increased risk of intracranial bleeding during delivery. They recommended that in the management of achondroplastic infants ultrasonography be done at birth and at 2, 4, and 6 months of age to establish ventricular size, the presence or absence of hydrocephalus, and possible intracranial bleed. They stated the impression that some achondroplasts have only megalencephaly, others have true communicating hydrocephalus, and yet others have dilated ventricles without hydrocephalus. Nelson et al. (1988) concluded that brainstem compression is common in achondroplasia and may account in part for the abnormal respiratory function.

Pauli et al. (1984) focused attention on the risk of sudden unexpected death in infants with achondroplasia. While uncontrolled and retrospective, their study demonstrated an excess of deaths in the first year of life, most or all of which were attributable to abnormalities at the craniocervical junction. Hecht et al. (1987) showed that the excess risk of death in infants with achondroplasia may approach 7.5%, largely because of cervical cord compression. Pauli et al. (1995) performed a prospective assessment of risk for cervical medullary-junction compression in 53 infants, 5 of whom were judged to have sufficient craniocervical junction compression to require surgical decompression. Intraoperative observation showed marked abnormality of the cervical spinal cord, and all operated-on children showed marked improvement of neurologic function. The best predictors of need for suboccipital decompression included lower-limb hyperreflexia or clonus on examination, central hypopnea demonstrated by polysomnography, and foramen magnum measures below the mean for children with achondroplasia.

Lachman (1997) reviewed the neurologic abnormalities in the skeletal dysplasias from a clinical and radiologic perspective. Three important major groups were identified: (i) achondroplasia (cranio-cervical junction problems in infancy, spinal stenosis, and neurogenic claudication in adulthood); (ii) type II collagenopathies (upper cervical spine anatomic and functional problems); and (iii) craniotubular and sclerosing bone dysplasias (osseous overgrowth with foraminal obstruction problems).

To detect myelopathy, Boor et al. (1999) recorded somatosensory evoked potentials (SEPs) after median nerve stimulation in 30 patients with achondroplasia. In addition to the conventional technique, they employed a noncephalic reference electrode recording the subcortical waveforms N13b and P13, generated near the craniocervical junction. The findings were correlated with the clinical status and MRI results. The sensitivities of the SEPs were 0.89 for cervical cord compression, 0.92 for myelomalacia, and 1.0 for the clinically symptomatic patients. There were no false-positive results. The subcortical SEPs were more sensitive than the conventional recordings.

Hecht et al. (1988) reviewed the subject of obesity in achondroplasia, concluding that it is a major problem which, whatever its underlying cause, aggravates the morbidity associated with lumbar stenosis and contributes to the nonspecific joint problems and to the possible early cardiovascular mortality in this condition. Using data about 409 Caucasian patients with achondroplasia from different countries (1,147 observations), Hunter et al. (1996) developed weight for height (W/H) curves for these patients. They showed that to a height of about 75 cm, the mean W/H curves are virtually identical for normal and achondroplastic children. After this height, the W/H curves for achondroplastic patients rise above those for the general population. Hunter et al. (1996) contended that the best estimation of weight excess for achondroplastic patients aged 3 to 6 years is given by the Quetelet index, whereas that for patients aged 6 to 18 years is the Rohrer index.

Homozygosity for the achondroplasia gene results in a severe disorder of the skeleton with radiologic changes qualitatively somewhat different from those of the usual heterozygous achondroplasia; early death results from respiratory embarrassment from the small thoracic cage and neurologic deficit from hydrocephalus (Hall et al., 1969). Yang et al. (1977) reported upper cervical myelopathy in a homozygote.

Horton et al. (1988) found that the epiphyseal and growth plate cartilages have a normal appearance histologically, and the major matrix constituents exhibit a normal distribution by immunostaining; however, morphometric investigations have indicated that the growth plate is shorter than normal and that the shortening is greater in homozygous than in heterozygous achondroplasia, suggesting a gene dosage effect. Stanescu et al. (1990) reported histochemical, immunohistochemical, electron microscopic, and biochemical studies on upper tibial cartilage from a case of homozygous achondroplasia. No specific abnormality was defined. Aterman et al. (1983) expressed puzzlement at the striking histologic changes in homozygous achondroplasia despite the virtual absence of changes in the heterozygote. They pointed out that histologic studies in the heterozygote at a few weeks or months of age have not been done. They suggested that because of similarities between what they called PHA (presumed homozygous achondroplasia) and thanatophoric dwarfism (187600), some cases of the latter condition may be due to a particularly severe mutation at the achondroplasia locus.

Young et al. (1992) described lethal short-limb dwarfism in the offspring of a father with spondyloepiphyseal dysplasia congenita (SEDC; 183900) and a mother with achondroplasia. Young et al. (1992) suggested that the infant was a double heterozygote for the 2 dominant genes rather than a compound heterozygote. It was considered unlikely that SEDC and achondroplasia are allelic because of the evidence that most, if not all, cases of SEDC result from mutation in the type II collagen gene (COL2A1; 120140), whereas this gene has been excluded as the site of the mutation in achondroplasia.

Evidence that hypochondroplasia (146000) can be caused by an allele at the achondroplasia locus came from observations of a presumed genetic compound in the offspring of an achondroplastic father and a hypochondroplastic mother who exhibited growth deficiency and radiographic abnormalities of the skeleton that were much more severe than those typically seen in achondroplasia (McKusick et al., 1973; Sommer et al., 1987) and somewhat less severe than those of the ACH homozygote. Huggins et al. (1999) reported an 8-month-old girl with achondroplasia/hypochondroplasia whose father had the G380R achondroplasia mutation (134934.0001) in the FGFR3 gene and whose mother had the N450K hypochondroplasia mutation (134934.0010). Chitayat et al. (1999) simultaneously reported an infant boy with achondroplasia/hypochondroplasia whose mother had the G380R mutation and whose father had the N450K mutation. Molecular analysis confirmed the compound heterozygosity of both children, who displayed an intermediate phenotype that was more severe than either condition in the heterozygous state but less severe than homozygous ACH.

In a presentation of adult genetic skeletal dysplasias found in the Museum of Pathological Anatomy in Vienna, Beighton et al. (1993) pictured the skeleton of a 61-year-old man with achondroplasia who died of transverse myelitis. Randolph et al. (1988) reported an achondroplastic patient who developed classic ankylosing spondylitis (106300). There is no fundamental connection between the 2 disorders. The importance of the observation is mainly to indicate that back problems in achondroplasts can be due to causes other than the underlying disease.

Hunter et al. (1998) presented data from a multicenter study of 193 individuals with achondroplasia. They found that 89.4% of children had at least one episode of otitis media within the first 2 years of life; 24 of 99 children who had otitis media in the first year of life had several infections. All were observed to have chronic otitis media; 78.3% of individuals required the insertion of ventilation tubes at some point in their lives. Thirty of 85 patients aged 1 to 2 years and 26 of 70 patients aged 2 to 3 years had received at least one set of ventilation tubes. A degree of conductive hearing loss was found in 38.3% of individuals at sometime in their lives, the majority of these being found after 4 years of age Tonsillectomy was performed in 38.8% of individuals, with cumulative rates of 8.8% within the first 4 years of life and 25% by age 8 years. Speech delay was found in 18.6% of individuals, and 10.9% had articulation problems; only 9.5% of these individuals received speech therapy. Orthodontic problems were found in 53.8% of individuals; only 3.2% of these individuals presented within the first 10 years of life.

Hunter et al. (1998) found that 10.5% of individuals had a ventricular shunt placed; all but one of these procedures were done in the preteenage years. Cervicomedullary decompression surgery had been performed in 6.8% of children by 4 years of age; however, this procedure was also performed in a number of older children, teenagers, and adults, with a total of 16.5% of individuals having this type of surgery. Apnea was reported in 10.9% of individuals by age 4 years and 16.1% of individuals overall.

Hunter et al. (1998) defined tibial bowing as a distance of greater than 5 cm between the knees, with the legs straight and ankles apposed. Using these criteria, they found that 9.7% of individuals had tibial bowing by age 5 years. This continued to develop throughout childhood and into adult life, with a total of 41.6% of individuals being affected at some time. Tibial osteotomy had been performed on 21.6% of these individuals. By age 10 years, 8.9% of individuals had neurologic signs in the leg; however, by the sixth decade, 77.9% of individuals had these signs. A total of 24.1% had surgery for spinal stenosis, with an additional 18% in whom the diagnosis was made but surgery had not been performed. A majority of these surgeries were performed in individuals over 40 years of age. Hunter et al. (1998) concluded that middle ear disease with its attendant risk of hearing loss was more frequent than previously reported, and that while a significant number of patients with achondroplasia experience delayed speech, only a minority receive speech therapy. The rate of early cervicomedullary decompression was comparable to the previously reported series, but an equivalent proportion of patients require such intervention beyond childhood. Hunter et al. (1998) also concluded that a significant number of patients have neurologic complaints by their teenage years and that this becomes a majority in adulthood.

Tasker et al. (1998) characterized cardiorespiratory and sleep dysfunction in 17 patients with achondroplasia referred to Great Ormond Street Hospital for Children, London. Three distinct etiologic groups were identified: group 1 had a mild degree of midfacial hypoplasia resulting in relative adenotonsillar hypertrophy; group 2 had jugular foramen stenosis resulting in muscular upper airway obstruction and progressive hydrocephalus due to jugular venous hypertension; and group 3 had muscular upper airway obstruction without hydrocephalus resulting from hypoglossal canal stenosis with or without foramen magnum compression. In addition, gastroesophageal reflux, which tended to occur in group 3 patients, was identified as a significant factor in the development of airway disease. Group 1 patients had obstructive sleep apnea only, and showed marked symptomatic improvement following adenotonsillectomy. Group 2 patients had central apnea responsive to surgical treatment of their hydrocephalus; obstructive sleep apnea in this group did not appear to respond to adenotonsillectomy, but to nocturnal continuous positive airway pressure. Group 3 patients had progressive cor pulmonale, obstructive and central sleep apnea, and gastroesophageal reflux with small airway pathology requiring multiple treatment modalities including foramen magnum decompression.

In 4 (3.2%) of 126 children with achondroplasia undergoing periodic evaluations at a bone dysplasia clinic, Pauli and Modaff (1999) identified a right-sided temporal bone abnormality involving absence of a roof over the jugular bulb, with bulging of the bulb into the middle ear cavity. In 2 patients, dark bluish-gray discoloration behind the tympanic membrane was noted, and temporal bone CT scan confirmed the presence of unilateral jugular bulb dehiscence. In a third patient, a large dehiscent jugular bulb was observed during exploratory tympanotomy; in a fourth patient, after brisk bleeding during attempted myringotomy and tube placement, CT scan demonstrated the absence of the bony covering of the jugular bulb. Jugular bulb dehiscence was suspected in a fifth patient with dark bluish discoloration behind the inferior quarter of the tympanic membrane, but confirmatory studies had not been performed at the time of the report. Pauli and Modaff (1999) noted that dehiscence of the jugular bulb is of clinical relevance, particularly in regard to difficult-to-control bleeding at myringotomy, and is associated with otherwise unexplained hearing loss, tinnitus, and self-audible bruits in children with achondroplasia.

Reynolds et al. (2001) retrospectively reviewed clinical and computed tomographic data in 71 infants with achondroplasia. They found no correlation between infantile hypotonia and foramen magnum size. These results suggested that there is no direct relationship and that foraminal size does not affect severity of hypotonia. They concluded that the only plausible explanation for the infantile hypotonia of achondroplasia is a primary effect of the causative mutation in FGFR3 (134934), which is expressed in brain.

Van Esch and Fryns (2004) described acanthosis nigricans in a 9-year-old boy with achondroplasia due to the classic gly380-to-arg mutation (134934.0001) in FGFR3.

Wynn et al. (2007) reported a 42-year follow-up study of mortality in achondroplasia. The study included 718 achondroplasia individuals from an earlier mortality study by Hecht et al. (1987) and 75 additional achondroplasia individuals. Rates of death were similar across the entire follow-up period. The overall mortality and age-specific mortality at all ages remained significantly increased. Accidental and neurologic disease-related deaths were increased in adults. Heart disease-related mortality, between ages 25 and 35, was more than 10 times higher than in the general population. Overall survival and the average life expectancy in this ACH population were decreased by 10 years.

Inheritance

Achondroplasia is inherited as an autosomal dominant with essentially complete penetrance. About seven-eighths of cases are the result of new mutation, there being a considerable reduction of effective reproductive fitness.

Paternal age effect on mutation was noted by Penrose (1955). Stoll et al. (1982) reported advanced paternal age in sporadic cases ascertained through the French counterpart of LPA (Little People of America), APPT (Association des Personnes de Petite Taille). Thompson et al. (1986) found that, on average, the severity of achondroplasia tends to be reduced with increasing parental age. It is doubtful that a recessive form of achondroplasia, indistinguishable from the dominant form, exists. Documentation of the diagnosis is inadequate in most reports of possible recessive inheritance.

Tiemann-Boege et al. (2002) performed direct molecular measurement of germline mutation frequency in sperm to test the hypothesis of paternal age effect on mutation. Using sperm DNA from donors of different ages, they determined the frequency of the 1138G-A mutation in the FGFR3 gene (134934.0001) that causes achondroplasia. The magnitude of the increase in mutation frequency with age was insufficient to explain why older fathers have a greater chance of having a child with this condition. A number of alternatives were considered to explain this discrepancy, including selection for sperm that carry the mutation or an age-dependent increase in premutagenic lesions that remain unrepaired in sperm and were inefficiently detected by the PCR assay used in the study.

Cohn and Weinberg (1956) reported affected twins with an affected sib. (This may have been achondrogenesis, e.g., 200600.) Chiari (1913) reported affected half sibs whose father had achondroplasia. Two first cousins, whose mothers were average-statured sisters, had undoubted achondroplasia (Wadia, 1969). Most dominants show sufficient variability to account for observations such as these on the basis of reduced penetrance but such is not the case with achondroplasia.

Gonadal mosaicism (or spermatogonial mutation) is a possible explanation for affected sibs from normal parents. Bowen (1974) described a possible instance of gonadal mosaicism; 2 daughters of normal parents had achondroplasia. One of the daughters had 2 children, one of whom was also achondroplastic. Fryns et al. (1983) reported 3 achondroplastic sisters born to normal parents. Philip et al. (1988) described the case of a man who had 3 daughters with classic achondroplasia, by 2 different women.

Henderson et al. (2000) reported sibs with achondroplasia born to average-statured parents. Both children had the 1138G-C causal mutation (134934.0002); this was also found in 28% of the unaffected mother's peripheral leukocytes. The authors therefore hypothesized that she was a germline as well as somatic mosaic for this mutation.

Sobetzko et al. (2000) also reported achondroplasia in a brother and sister with unaffected parents. The sibs shared the classic 1138G-A mutation (134934.0001) and also shared a 4p haplotype derived from the unaffected father. Paternal sperm was not available, and evidence of gonadal mosaicism could not be substantiated.

Affected cousins could be due to the coincidence of 2 independent mutations. Such was probably the case, in McKusick's opinion, in the second cousins once removed reported by Fitzsimmons (1985). Reiser et al. (1984) reviewed 6 families with unexpected familial recurrence and hypothesized that these recurrences were simply the result of 2 independent chance events. Dodinval and Le Marec (1987) reported 2 families, each with 2 cases of achondroplasia. In 1 family, a girl and her great aunt were affected; in the other, male and female first cousins. Both germinal mosaicism and paternal age effect appear to have their basis in the way spermatogonia are replenished, a feature that distinguishes gametogenesis in the male from that in the female. As outlined by Clermont (1966), spermatogonia go through a few mitotic divisions before embarking on the meiotic divisions that lead to mature sperm. Some of the products of the mitotic divisions are returned to the 'cell bank' to replenish the supply of spermatogonia. Mutations occurring during DNA replication can, therefore, accumulate, providing a basis for paternal age effect and for germinal mosaicism. Hoo (1984) suggested a small insertional translocation as a possible mechanism for recurrent achondroplasia in sibs with normal parents. In discussing 'male-driven evolution' and the evidence for a generally higher mutation rate in males than in females, Crow (1997) stated that the number of cell divisions required to generate sperm cells in a 30-year-old man is estimated at 400; the number of cell divisions that generate an egg is 24, irrespective of age. If mutation rates are proportional to the cell divisions, the male-to-female ratio should equal 17. In fact, the data show a higher ratio, as if mutation rates increase at a higher rate than the number of replications would predict--not surprising if fidelity of transcription and efficiency of repair mechanisms diminish with age. Studies in male and female birds by Ellegren and Fridolfsson (1997) appeared to support male-driven evolution of DNA sequences in birds.

The severe phenotype of the homozygote for the ACH gene and the possibility that hypochondroplasia represents an allelic disorder were discussed in connection with the discussion of clinical features of achondroplasia.

Langer et al. (1993) described a patient who was doubly heterozygous for achondroplasia and pseudoachondroplasia (177170). Woods et al. (1994) described a family in which the father had pseudoachondroplasia and the mother had achondroplasia, and 2 daughters were doubly affected and a son had achondroplasia only. At birth, the 2 daughters appeared to have achondroplasia. Later, the development of a fixed lumbar gibbus, unusual radiographic changes in the spine, increasing joint laxity of the hands, and characteristic gait and hand posture made the appearance of pseudoachondroplasia apparent.

Flynn and Pauli (2003) described a fourth case with radiologic findings virtually identical to those described by Langer et al. (1993) and Woods et al. (1994). They commented that the fact that all the probands were initially thought to have achondroplasia alone is not surprising, since pseudoachondroplastic features usually are not identifiable until after 2 years of age in affected individuals. The patient described by Langer et al. (1993) developed lumbar spinal stenosis at age 7.5 years. Both sibs in the report of Woods et al. (1994) had sufficiently severe stenosis of the foramen magnum to cause high cervical myelopathy requiring decompression.

Flynn and Pauli (2003) described a family in which the proband, her mother, and her maternal grandfather were all double heterozygotes for achondroplasia and for osteogenesis imperfecta type I (166200). Radiographic and clinical examination demonstrated features of both conditions, with neither being more prominent than would be expected for an individual heterozygous for each disorder alone.

Because of gonadal mosaicism, the risk of recurrence of achondroplasia in the sibs of achondroplastic children with unaffected parents is presumably higher than twice the mutation rate, but had not been measured. Mettler and Fraser (2000) collected data from 11 Canadian genetics centers and arrived at an estimate of 1 in 443, or 0.02%.

Stoll and Feingold (2004) performed analyses to determine whether a connection between teratogenesis and carcinogenesis is indicated by a higher cancer risk in parents of children with congenital anomalies. In achondroplasia, the new mutations are of paternal origin, raising the hypothesis of the existence of a 'mutator' gene acting in male meiosis and in somatic, mitotic cells in both sexes, which may favor the occurrence of cancer. By a questionnaire survey involving 76 males and 72 females with achondroplasia, Stoll and Feingold (2004) found that paternal grandfathers and grandmothers had significantly more cancers (56) than maternal grandfathers and grandmothers (24) (chi square = 14.80, p less than 0.001).

In 3 sibs who were the product of the first and third pregnancies of healthy nonconsanguineous parents, Natacci et al. (2008) identified heterozygosity for the G380R mutation in the FGFR3 gene (134934.0001). The mutation was not found in lymphocytic DNA from the parents; however, DNA analysis of a sperm sample from the 37-year-old father showed the G380R mutation. The authors stated that this was the second reported case of germinal mosaicism causing recurrent achondroplasia in a subsequent conception.

Mapping

Strom (1984) and Eng et al. (1985) purported to find abnormality of the type II collagen gene in achondroplasia. If such a defect is present, one might expect ocular abnormality in achondroplasia inasmuch as type II collagen is present in vitreous. SED congenita was a more plausible candidate for a structural defect of type II collagen because it is a dominant disorder that combines skeletal dysplasia with vitreous degeneration and deafness (experimental studies with antibodies to type II collagen indicate that this collagen type is represented in the middle ear); subsequently, defects were in fact found in the COL2A1 gene in SEDC. The report by Eng et al. (1985) was withdrawn in 1986 because figures, 'which were generated in the laboratory of C. Strom and C. Eng, were improperly assembled and therefore cannot be used to support the conclusions of the article.' Francomano and Pyeritz (1988) excluded COL2A1 as the site of the mutation in achondroplasia by use of probes spanning the gene in an analysis of genomic DNA from 49 affected persons and 2 multiplex families. No gross rearrangements were seen on Southern blot analysis, and linkage studies in the multiplex families demonstrated discordant inheritance of achondroplasia and COL2A1 alleles. Evidence against linkage to COL2A1 has been presented before by Ogilvie et al. (1986). From their studies, Finkelstein et al. (1991) concluded that mutations at the chondroitin sulfate proteoglycan core protein (CSPGP) locus do not cause achondroplasia or pseudoachondroplasia (177170).

Edwards et al. (1988) commented on a report, made at the national meeting of the Neurofibromatosis Foundation, of 2 individuals with achondroplasia and neurofibromatosis (162200) who had translocations involving the long arm of chromosome 17. In both cases the breakpoint was at the region consistent with localization of the neurofibromatosis gene by linkage studies; a third case of coincident achondroplasia and neurofibromatosis was also mentioned. Korenberg et al. (1989) and Pulst et al. (1990) demonstrated by linkage analysis that the achondroplasia locus does not map between the 2 groups of markers flanking the gene for neurofibromatosis-1 on human chromosome 17. Verloes et al. (1991) observed connatal neuroblastoma in an infant with achondroplasia and suggested that the achondroplasia gene may be located on the short arm of chromosome 1 where a neuroblastoma locus (see 256700) appears to be situated.

By linkage studies using DNA markers, Velinov et al. (1994) and Le Merrer et al. (1994) mapped the gene for achondroplasia and hypochondroplasia to the distal area of the short arm of chromosome 4 (4p16.3). Francomano et al. (1994) likewise mapped the ACH gene to 4p16.3, using 18 multigenerational families with achondroplasia and 8 anonymous dinucleotide repeat polymorphic markers from this region. No evidence of genetic heterogeneity was found. Analysis of a recombinant family localized the ACH locus to the 2.5-Mb region between D4S43 and the telomere.

Molecular Genetics

Once the gene for achondroplasia was assigned to 4p16.3 by linkage analysis (Le Merrer et al., 1994; Velinov et al., 1994; Francomano et al., 1994), causative mutations were identified by the candidate gene approach and reported within 6 months of the first mapping report. Mutations in the gene for fibroblast growth factor receptor-3 (134934) were identified by Shiang et al. (1994) and independently by Rousseau et al. (1994). The FGFR3 gene had previously been mapped to the same region, 4p16.3, as the ACH gene and the Huntington disease gene. The mutation in 15 of the 16 achondroplasia-affected chromosomes studied by Shiang et al. (1994) was the same, a G-to-A transition at nucleotide 1138 (134934.0001) of the cDNA. The mutation on the only other ACH-affected chromosome 4 without the G-to-A transition at nucleotide 1138 had a G-to-C transversion at this same position (134934.0002). Both mutations resulted in the substitution of an arginine residue for a glycine at position 380 of the mature protein, which is in the transmembrane domain of FGFR3. The mutation was located in a CpG dinucleotide. Rousseau et al. (1994) found the G380R mutation in all cases studied: 17 sporadic cases and 6 unrelated familial cases. Because of the high mutation rate, it might have been predicted that the achondroplasia gene is large and that any one of many mutations could lead to the same or a similar (hypochondroplasia) phenotype. Such is apparently not the case. The fact that there are no reports of Wolf-Hirschhorn syndrome (194190) patients with stigmata of achondroplasia may indicate that the phenotype is due to some mechanism other than haploinsufficiency, e.g., represents a dominant-negative or gain-of-function effect. (The independent work of Shiang et al. (1994) and Rousseau et al. (1994) was reported in the 29 July issue of Cell and the 15 September issue of Nature, respectively.)

Bellus et al. (1995) found that 150 of 154 unrelated achondroplasts had the G-to-A transition (134934.0001) and 3 had the G-to-C transversion (134934.0002) at nucleotide 1138 of the FGFR3 gene. All 153 had the gly380-to-arg substitution; in one individual, an atypical case, the gly380-to-arg substitution was missing. Nucleotide 1138 of the FGFR3 gene was the most mutable nucleotide in the human genome discovered at that time. Superti-Furga et al. (1995) reported the case of a newborn with achondroplasia who did not carry the mutation at nucleotide 1138 changing glycine-380 to arginine but had a mutation causing substitution of a nearby glycine with a cysteine (134934.0003).

The FGFR3 gene was isolated and studied in connection with a search for the Huntington disease gene. The distribution of FGFR3 mRNA in embryonic mouse tissues was found to be more restricted than that of FGFR1 (136350) and FGFR2 (176943) mRNA. Outside of the developing central nervous system, the highest level of FGFR3 mRNA was found to be in the prebone cartilage rudiments of all bones, and during endochondral ossification, FGFR3 was detected in resting but not hypertrophic cartilage (Peters et al., 1993). The glycine-to-arginine substitution would have a major effect on the structure, function, or both of the hydrophobic transmembrane domain and most likely would have a significant effect on the function of the receptor. Five of 6 ACH homozygotes were homozygous for the G-to-A transition and each of 6 sporadic cases, including the parents of 2 of the homozygotes, were heterozygous for the 1138A allele and the wildtype allele. The fact that FGFR3 transcripts are present in fetal and adult brain (which has the highest levels of any tissue) may have relevance in connection with the megalencephaly which is thought to occur in achondroplasia (Dennis et al., 1961).

FGFR3 codes for at least 2 isoforms of the gene product by alternate use of 2 different exons that encode the last half of the third immunoglobulin domain (IgIII), which is primarily responsible for the ligand-binding specificity. The isoforms are preferentially activated by the various fibroblast growth factors.

Rump et al. (2006) reported a Dutch infant with a severe form of achondroplasia caused by 2 de novo mutations in the FGFR3 gene on the same allele: the common G380R mutation (134934.0001) and L377R (134934.0027). Allele-specific PCR analysis confirmed that the 2 mutations were in cis. From birth, the child had severe respiratory difficulties with multiple hypoxic episodes due to a combination of upper airway obstruction, pulmonary hypoplasia, and cervicomedullary compression. He eventually became ventilator dependent and died at age 4 months.

Horton (2006) reviewed work on the nature of the basic defect in achondroplasia. After mutations in FGFR3 were identified as the basis of achondroplasia in 1994, attention turned to how the mutation disturbed linear bone growth. Biochemical studies of the FGFR3 receptor combined with knockout experiments in mice revealed that FGFR3 is a negative regulator of chondrocyte proliferation and differentiation in the growth plate and that the mutations in achondroplasia and related disorders activate the receptor. Thus they can be viewed as gain-of-function mutations.

Heuertz et al. (2006) screened 18 exons of the FGFR3 gene in 25 patients with hypochondroplasia and 1 with achondroplasia in whom the common mutations G380R and N540K had been excluded. The authors identified 7 novel missense mutations, including 1 in the patient with achondroplasia (S279C; 134934.0030). Heuertz et al. (2006) noted that 4 of the 6 extracellular mutations created additional cysteine residues and were associated with severe phenotypes.

Diagnosis

The diagnosis is based on the typical clinical and radiologic features; the delineation from severe hypochondroplasia may be arbitrary.

The demonstration of a very limited number of mutations causing achondroplasia and the ease with which they can be detected (1 PCR and 1 restriction digest) provides a simple method for prenatal diagnosis of ACH homozygotes in families at risk and in which the parents are heterozygous for either the 1138A or 1138C allele (Shiang et al., 1994). Shiang et al. (1994) expressed the opinion that other than the screening of at-risk pregnancies for homozygous ACH fetuses, any 'other application of the diagnostic test for ACH mutations should be prohibited.' Bellus et al. (1994) practiced prenatal diagnosis by chorionic villus sampling at 10 weeks and 4 days of gestation, both parents having achondroplasia. Both parents and the fetus were shown to be heterozygous for the more common G-to-A transition. Homozygous achondroplasia was excluded.

Clinical Management

Recommendations for follow-up and management were reviewed at the first international symposium on achondroplasia (Nicoletti et al., 1988) and by Horton and Hecht (1993). The recommendations included: measurements of growth and head circumference using growth curves standardized for achondroplasia (Horton et al., 1978); careful neurologic examinations (including CT, MRI, somatosensory evoked potentials and polysomnography) and surgical enlargement of the foramen magnum in cases of severe stenosis; management of frequent middle ear infections and dental crowding; measures to control obesity starting in early childhood; growth hormone therapy (Horton et al., 1992), which is still experimental, and lengthening of the limb bones; tibial osteotomy or epiphysiodesis of the fibular growth plate to correct bowing of the legs; lumbar laminectomy for spinal stenosis which typically manifests in early adulthood; delivery of pregnant women with achondroplasia by cesarean section; and prenatal detection of affected fetuses by ultrasound.

Hunter et al. (1996) recommended that achondroplastic children stay within 1 SD of the mean weight for height curves for achondroplasts.

Hoover-Fong et al. (2007) developed weight for age, gender-specific growth curves for children with achondroplasia from birth through 16 years. The charts were constructed from a longitudinal, retrospective, single observer cohort study of 334 individuals with achondroplasia. The investigators proposed that the charts could be used in conjunction with current height for age charts developed by Horton et al. (1978) and weight for height charts developed by Hunter et al. (1996).

Shohat et al. (1996) investigated the effect of recombinant human growth hormone (hGH) treatment on the growth rate and proportion of individuals with achondroplasia and hypochondroplasia. They studied 15 individuals over 24 months including 6 months of observation, 12 months of hGH therapy (0.04 mg/kg/day), and 6 months of post treatment growth rate determination. The mean growth rate during hGH treatment (5.3 +/- 1.6 cm) of achondroplasts was significantly increased compared to pretreatment (4.0 +/- 1.0 cm/year, P less than 0.01) and posttreatment periods (3.1 +/- 1.3 cm; P less than 0.001). In the 4 children with hypochondroplasia, the growth rate during hGH treatment was 7.0 +/- 2.4 cm/year and 4.9 +/- 1.5 cm/year during the pre- and posttreatment periods, respectively. In achondroplasts, there was a significant increase in growth rate of only the lower segment (from 1.1 +/- 1.6 cm/year to 3.1 +/- 1.2 cm/year, P less than 0.02). Unexpectedly, this treatment does not seem to have a lesser effect on limbs than on trunk growth rate and, therefore, during 1 year of treatment, does not increase body disproportion.

Waters et al. (1995) studied the results of treatment of obstructive sleep apnea in achondroplasia. Treatment included adenotonsillectomy, weight loss, and nasal-mask continuous positive airway pressure (CPAP). They observed improvements in measurements of disturbed sleep architecture and some evidence of improvement in neurologic function.

Weber et al. (1996) studied the effects of recombinant human growth hormone treatment in 6 prepubertal children with achondroplasia, ranging in age from 2 to 8 years. They were given a GH dose of 0.1 IU/kg/day subcutaneously. During the year of treatment the growth velocity increased from 1.1 to 2.6 cm/year in 3 patients while in the others no variation was detected. No side effects were observed during the trial apart from the slight advancement of bone age in 2 patients. Their findings confirmed the individual variability in the response to GH treatment.

Horton (2006) reviewed milestones in achondroplasia research. As the molecular pathogenesis of achondroplasia emerged, interest shifted to therapy intended to counter the effects of the overactive receptor. One strategy involved chemical inhibitors selected for the FGFR3 tyrosine kinase. A second relied on blocking antibodies to interfere with binding of FGF ligands to FGFR3 (Aviezer et al., 2003). A third possibility involved C-type natriuretic peptide (CNP; 600296) which had been shown by Yasoda et al. (2004) to downregulate FGF-induced activation of MAP kinase signaling pathways in growth plate chondrocytes and to counteract the effects of the achondroplasia mutation in mice.

In achondroplasia and thanatophoric dysplasia (187600), spinal canal and foramen magnum stenosis can cause serious neurologic complications. Matsushita et al. (2009) observed premature synchondrosis closure in the spine and cranial base in human cases of homozygous achondroplasia and thanatophoric dysplasia as well as in mouse models of achondroplasia. In both species, premature synchondrosis closure was associated with increased bone formation. Chondrocyte-specific activation of Fgfr3 in mice induced premature synchondrosis closure and enhanced osteoblast differentiation around synchondroses. FGF signaling in chondrocytes increased bone morphogenetic protein (Bmp) ligand (e.g., BMP7, 112267) mRNA expression and decreased Bmp antagonist (e.g., noggin, 602991) mRNA expression in a MAPK-dependent manner, suggesting a role for Bmp signaling in the increased bone formation. The enhanced bone formation would accelerate the fusion of ossification centers and limit the endochondral bone growth. The authors proposed that spinal canal and foramen magnum stenosis in heterozygous achondroplasia patients may occur through premature synchondrosis closure. If this is the case, then any growth-promoting treatment for these complications of achondroplasia must precede the timing of the synchondrosis closure.

C-type natriuretic peptide (CNP) antagonizes FGFR3 downstream signaling by inhibiting the pathway of mitogen-activated protein kinase (MAPK). Lorget et al. (2012) reported the pharmacologic activity of a 39-amino acid CNP analog (BMN 111) with an extended plasma half-life due to its resistance to neutral endopeptidase (NEP; 120520) digestion. In achondroplasia human growth plate chondrocytes, Lorget et al. (2012) demonstrated a decrease in the phosphorylation of extracellular signal-regulated kinases 1 (ERK1; 601795) and 2 (ERK2; 176948), confirming that this CNP analog inhibits FGF-mediated MAP kinase activation. Concomitantly, Lorget et al. (2012) analyzed the phenotype of Fgfr3(Y367C/+) mice and showed the presence of achondroplasia-related clinical features in this mouse model. Lorget et al. (2012) found that in Fgfr(Y367C) heterozygous mice, treatment with the CNP analog led to a significant recovery of bone growth. They also observed an increase in the axial and appendicular skeleton lengths and improvements in dwarfism-related clinical features including flattening of the skull, reduced crossbite, straightening of the tibias and femurs, and correction of the growth plate defect. Lorget et al. (2012) concluded that their results provided the proof of concept that BMN 111, a NEP-resistant CNP analog, might benefit individuals with achondroplasia and hypochondroplasia.

Savarirayan et al. (2019) reported the results of a phase 2 dose-finding and extension study of vosoritide (a biologic analog of C-type natriuretic peptide) given by once-daily subcutaneous injection in 35 children with achondroplasia aged 5 through 14 years. All patients had adverse events (most commonly injection-site reactions), and serious adverse events occurred in 4 of the 35 patients. Therapy was discontinued in 6 patients, in 1 due to an adverse event. During the first 6 months of treatment, a dose-dependent increase in the annualized growth velocity was observed up to a dose of 15 mcg/kg, and a sustained increase was observed at doses of 15 and 30 mcg/kg for up to 42 months. There was no difference in efficacy or safety between the 15 and 30 mcg/kg doses, which supported the choice of the lower dose for further evaluations.

Population Genetics

Early estimates on the prevalence of achondroplasia are undoubtedly incorrect because of misdiagnosis. For example, Wallace et al. (1970) reported 2 female sibs as examples of achondroplasia; both died in the neonatal period and showed, in addition to chondrodystrophy, central harelip, hypoplastic lungs, and hydrocephalus. Without radiographic studies it is impossible to identify the nature of this condition, but it is certainly not true achondroplasia; Jeune asphyxiating thoracic dystrophy (208500), thanatophoric dwarfism (187600), and achondrogenesis are each possibilities.

Using modern diagnostic criteria, Gardner (1977) estimated the mutation rate at 0.000014. Orioli et al. (1986) reported on the frequency of skeletal dysplasias among 349,470 births (live and stillbirths). The prevalence rate for achondroplasia was between 0.5 and 1.5/10,000 births. The mutation rate was estimated to be between 1.72 and 5.57 x 10(-5) per gamete per generation. The stated range is a consequence of the uncertainty of diagnosis in some cases. (The thanatophoric dysplasia/achondrogenesis group had a prevalence between 0.2 and 0.5/10,000 births. Osteogenesis imperfecta had a prevalence of 0.4/10,000 births. Only 1 case of diastrophic dysplasia was identified.) In the county of Fyn in Denmark, Andersen and Hauge (1989) determined the prevalence of generalized bone dysplasias by study of all children born in a 14-year period. The figures, which