Fibroblast Growth Factor Receptor 2 (FGFR2) Mutation Related Syndromic Craniosynostosis

Abstract

Craniosynostosis results from the premature fusion of cranial sutures, with an incidence of 1 in 2,100-2,500 live births. The majority of cases are non-syndromic and involve single suture fusion, whereas syndromic cases often involve complex multiple suture fusion. The fibroblast growth factor receptor 2 (FGFR2) gene is perhaps the most extensively studied gene that is mutated in various craniosynostotic syndromes including Crouzon, Apert, Pfeiffer, Antley-Bixler, Beare-Stevenson cutis gyrata, Jackson-Weiss, Bent Bone Dysplasia, and Seathre-Chotzen-like syndromes. The majority of these mutations are missense mutations that result in constitutive activation of the receptor and downstream molecular pathways. Treatment involves a multidisciplinary approach with ultimate surgical fixation of the cranial deformity to prevent further sequelae. Understanding the molecular mechanisms has allowed for the investigation of different therapeutic agents that can potentially be used to prevent the disorders. Further research efforts are need to better understand screening and effective methods of early intervention and prevention. Herein, the authors provide a comprehensive update on FGFR2-related syndromic craniosynostosis.

Introduction

Craniosynostosis results from the premature fusion of cranial sutures, with an incidence of 1 in 2,100-2,500 live births ^1-4. In normal development, cranial growth occurs perpendicular to suture lines leading to a well-proportioned head with a longer anteroposterior than bitemporal dimension ⁵. Sutures typically fuse in an orderly fashion with the first suture - metopic - closing within the first year of life and the final sutures squamosal and lambdoidal not closing until the fourth or fifth decade of life. Premature fusion of one or more sutures precludes growth of cranial bones perpendicular to the affected sutures and leads to compensatory growth in a parallel direction, resulting in a cranial dysmorphic head shape ^6,7.

The vast majority of craniosynostoses involve single suture fusion, though multiple sutures can be affected ⁵. Depending on which sutures prematurely fused, patients will have characteristic phenotypes. There are four major sutures including sagittal, coronal, metopic and labdoid, as well as minor sutures including frontonasal, temporosquamosal, and frontosphenoidal. The sagittal suture is involved in 40-58% of cases, followed by unilateral or bilateral coronal (20-30%), metopic (5-15%) and lambdoid (2-4% of cases) suture fusion ^{3, 5, 8}. Sagittal suture fusion results in an increase of the anterior-posterior diameter resulting in dolichocephaly or scaphocephaly ⁵. Unilateral coronal synostosis results in anterior plagiocephaly whereas bilateral coronal fusion results in brachycephaly. Premature metopic suture fusion is associated with triangular shaped forehead, trigonocephaly. Lambdoid synostosis is usually unilateral, resulting in asymmetric posterior plagiocephaly, whereas bilateral lambdoid synostosis results in flattened and widened occipital region ⁵. Kleeblattschadel or cloverleaf skull is a term given for a trilobed skull resulting from synostosis of lambdoidal and coronal sutures.

Cases of craniosynostoses are thought to result from a complex interaction of environment, epigenetics, and genetics ^{9, 10}. Evidence from a population-based study suggests that multiple births, maternal age greater than 35, birth weight >4000 g, and male sex are risk factors for craniosynostosis ¹¹. Other potential risk factors include advanced paternal age, higher parental education, maternal exposure to valproate acid, preterm delivery, and in-utero growth constraint ^1,12-14. The vast majority of these cases are non-syndromic and 80-90% of cases involve a single suture line ¹⁵. On the contrary, approximately 10-20% of craniosynostosis cases are syndromic, mostly with autosomal dominant inheritance, incomplete penetrance, and variable expressivity ⁶. The syndromic synostoses, such as Crouzon, Apert, and Pfeiffer, may involve multiple sutures and extracranial phenotypes are mostly of the hands and feet ¹⁶. Occasionally complex or multisuture craniosynostosis occurs in patients without an identifiable syndrome ¹⁷. Although there have been reports of prenatal diagnoses of craniosynostosis, the majority are diagnosed postnatally including those associated with syndromes.

Recently, whole exome and genome sequencing has enabled identification of hundreds of mutations implicated in craniosynostosis. Genes most commonly mutated in familial craniosynostosis include FGFR2, FGFR3, twist homologue 1 (TWIST1) and ephrin-B1 (EFNB1) ^2,18-21 Of these, FGFR variants have been most extensively characterized and serve as a basis for understanding the molecular underpinnings of craniosynostosis. Herein, the authors provide a comprehensive update on FGFR2-related syndromic craniosynostosis.

Fibroblast growth factor (FGF) receptors (FGFRs)

The FGF family, comprised of at least 22 known FGF ligands, binds to five related receptor tyrosine kinases, resulting in receptor monomer dimerization, activation of kinase domains and phosphorylation of the receptor ^22-24. It is well known that FGF/FGFR signaling plays important roles in numerous biological processes, including bone development and homeostasis by controlling differentiation of mesenchymal and neuroectoermal cells ^25-27. Mutations in three members of the FGFR family (FGFR1, FGFR2, and FGFR3) result in unregulated FGF signaling and premature suture closure. Much of the current knowledge regarding the role of FGF/FGFR in cranial growth is derived from mouse models. With the exception of FGFR5, members of the FGFR gene family encode a receptor tyrosine kinase (RTK), which has an extracellular ligand-binding portion composed of immunoglobulin-like domains (IgI, IgII, and IgIII), a transmembrane region, and intracellular tyrosine kinase domains (TK1 and TK2) (Figure 1) ^{3, 28-32}.

Figure 1.

Schematic representation of FGFR signaling pathway. A typical FGFR gene encodes a tyrosine kinase (TK) receptor. This receptor has an extracellular ligand-binding sites, which are comprised of immunoglobulin-like domains (IgI, IgII and IgIII), transmembrane region and divided intracellular tyrosine kinase domains, TK1 and TK2. The signaling pathway mainly operates through at least 3 distinct pathways. Initiation of RAS/MAPK pathways starts with formation of FRS2 complex and regulates cell proliferation and differentiation. The PI3K/AKT pathway controls cells survival and fate determination after getting initiated by FRS2 complex formation. Activation of PKC pathway started with binding of PLCγ to the activated FGFR and regulates morphogenesis and migration of cells.

Within the IgII domain, a positively charged conserved region serves as the binding site for heparin sulphate ²⁶. FGF binding occurs at IgII and IgIII as well as the link between these Ig-like domains. Alternative splicing in the IgIII domain results in various isoforms of the receptor with differing affinities for FGF ligands. The N-terminal part of the IgIII domain is encoded by exon IIIa and the C-terminal part is encoded by either exon IIIb or IIIc ²⁶.

Ligand binding activates downstream pathways including phosphoinositide 3-kinase (PI3K)/Akt, protein kinase C, phospholipase C, and mitogen-activated protein kinase [20, 27, 28]. Downstream of RAS, the extracellular signal-related kinases 1 and 2 (ERK1/2) are key proteins mediating mitogen-activated protein kinase (MAPK) in FGFR2 signaling ^31,33. Recent evidence suggests that reduced levels of ERF, which encodes an inhibitory ETS transcription factor directly bound to ERK1/2, results in complex multisuture craniosynostosis in humans and mice ^32,34. This suggests that FGFR2 mediated ERK-MAPK activity regulates coronal suture formation ³⁴.

Craniosynostotic disorders caused by mutations in FGFR2 gene are perhaps the most extensively studied ³⁵. Both minor and major sutures fusion occurs in most patients with FGFR2-related craniofacial dyostosis ³⁶. Abundant FGFR2 expression is found in the cartilage of the cranial base ^25,37-39. FGFR2 in cranial sutures is largely expressed in differentiating osteoblasts and osteoprogenitor cells. Via downstream pathways, FGF/FGFR2 signaling mediates osteoprogenitor proliferation, differentiation, and apoptosis ^25,40. Early differentiation is thought to be a leading factor involved in premature suture fusion ³. As will be discussed, autosomal dominant FGFR2 mutations enhance receptor signaling through dimerization independent of ligand binding with resultant constitutive activation ^41,42. Most pathogenic variants of FGFR2 are missense variants, with less frequent reports of deletions, insertions, and splice site mutations ²⁸. Mutations in FGFR2 less commonly result in non-syndromic craniosynostoses. The discussion in this review will be limited to FGFR2-related syndromic craniosynostosis.

FGFR2-related syndromic craniosynostosis

Heterozygous activating mutations of FGFR2 result in several craniosynostosis syndromes, such as Crouzon, Apert, Pfeiffer, Antley-Bixler, Beare-Stevenson cutis gyrata, Jackson-Weiss, Bent Bone Dysplasia, and Seathre-Chotzen-like syndromes ^43-47 (Table 1). Whereas the majority of mutations in FGFR2 localize to two exons encoding the extracellular IgIII domain (IIIa/exon 8, and IIIc/exon 10), mutations have been identified in exons coding the tyrosine kinase region (exons 3, 5, 14, 15, 16 and 17), and exon 11 that encodes the transmembrane domain, as well as the IgI and IgII domains ^48-51 (Table 1).

Table 1.

FGFR2- Associated Craniosynostotic Syndromes

Syndrome	Inheritance	Mutations	Craniosynostosis Findings	Extracranial Phenotypes	Ref.
Crouzon Syndrome	Autosomal dominant	Cys278Phe, Trp289Gly, Tyr290Gly, Ser267Pro, Tyr328Cys, Gly338Arg, Tyr340His, Cys342Tyr, Cys342Arg, Cys342Phe, Cys342Ser, Cys342Trp, Ala344Gly, Asn549Thr, Ser347Cys, Ser354Cys	Bicoronal synostosis, pansynostosis late	Crouzonoid face (flattened forehead, proptosis, hypertelorism, beaked nose, midface hypoplasia), normal hands	[35, 51,54-56,61]
Pfeiffer Syndrome	Autosomal dominant	Ala314Ser, Asp321Ala, Thr342Pro, Cys278Phe, Cys342Tyr, Trp290Cys, Tyr340Cys, Cys342Tyr, Cys342Arg, Cys342Ser, Cys342Trp, Ser351Arg, Val359Phe	Type I: Bicoronal synostosisType II: Kleeblattschädel (cloverleaf skull) and extreme proptosisType III: Bicoronal synostosis, No cloverleaf skull	Nasal abnormalities, proptosis, hypertelorism, frontal bossing, broad thumbs and great toes, hydrocephalus with type II	[35,52, 55-57,61, 104]
Apert Syndrome	Autosomal dominant	Ser252Trp, Pro253Arg, Ser252Phe	Bicoronal synostosis	Midface hypoplasia, macrocephaly, downslated palpebral fissues, radial deviation of thumb, severe syndactyly of hands and feet	[3, 63-65]
Antley-Bixler Syndrome	Autosomal recessive/ dominant forms	Ser351CysTyr290Cys	Multisuture synostosis	Midface hypoplasia, radiohumeral synostosis, joint contractures and arachnodactyly	[77-82]
Beare-Stevenson cutis gyrata syndrome	Autosomal dominant	Tyr375Cys, Ser372Cys	Kleeblattschädel deformity	Cutis gyrata (furrowed skin), ear defects, anogenital deformities	[83-94]
Jackson-Weiss Syndrome	Autosomal dominant	Ala344Gly, Glu289Pro, Ser347Cys, Cys342Arg,Cys342Ser	Variable	Facial dysmorphism and foot malformations (hands typically normal)	[10, 35, 95-99]
Bent Bone Dysplasia syndrome	Autosomal dominant	Tyr381Asp, Met391Arg	Bicoronal synostosis	Dysmorphic oral maxillofacial features as well as hypoplastic pubis, clavicles, osteopenia, and bent long bones	[40, 99,100]
Seathre-Chotzen-like syndrome	Autosomal dominant	Gln289ProVV269-70del	Heterogeneous patterns of craniosynostosis: uni- or bicoronal, sagittal, or metopic synostosis	Syndactyly of hands and feet, ptosis, hearing loss, hypertelorism, and spine abnormalities	[35,101-103]

Both Crouzon and Pfieffer are typically inherited in an autosomal dominant fashion and have similar molecular mechanisms. Crouzon is usually the mildest of the syndromes. Bicoronal synostosis is most common in Crouzon syndrome; however, pansynostosis may also be a late presenting feature in some cases. The crouzonoid facial appearance, which consists of flattened forehead, proptosis, hypertelorism, beaked nose, and midface hypoplasia without major limb deformities characterizes the disorder ^{26, 52} (Table 1).

The flattened forehead is secondary to bicoronal synostosis. In Pfieffer syndrome, the craniofacial severity is variable and extracranial manifestations include big toes and broad radially deviated thumbs ⁵³. Various clinical subtypes of Pfieffer syndrome have been described, type 1 being compatible with life, and types 2 and 3 associated with more severe complex synostosis and higher associated mortality ⁵⁴. The craniofacial features of the Pfieffer syndromes differ somewhat: type 1 typically has moderate-to-severe midface hypoplasia, type 2 has cloverleaf skull and extreme proptosis, and type 3 is characterized by turribrachycephaly (a high and prominent forehead) as well as proptosis ²⁸. The FGFR2 mutations in Crouzon and Pfeiffer syndrome overlap, most often occurring in extracellular IgIII domain encoded by exons 8 (IIIa) or 10 (IIIc) ⁴⁶. Both syndromes may be caused by the Cys278Phe and Cys342Tyr missense mutations ^55-57. Severe cases of Pfeiffer syndrome are caused by a small subset of substitutions encoding Trp290Cys, Tyr340Cys, Cys342Arg or Ser351Cys ^58-60. The Trp290Cys mutation has also been linked to tracheal cartilaginous sleeves in children with syndromic craniosynostosis ⁶¹. Recently, Ke and colleagues presented a unique report on a young boy with Crouzon syndrome who had bicoronal craniosynostosis, proptosis, and maxillary hypoplasia and identified a heterozygous Ser267Pro mutation in exon 8 ⁵⁵. Other mutations identified in patients with features resembling the Crouzon/Pfeiffer syndrome include Leu617Phe mutation in the TK domain and Gly271Val in the IgIIIa domain of FGFR2 ⁶². A unique mutation leading to Ala315Ser in FGFR2 has also been identified in in a family with a mild phenotype of Crouzon craniofacial dyostosis but without overt craniosynostosis ⁶³. A report from Japan also identified a novel FGFR2 Asn549Thr mutation in a patient with Crouzon syndrome ¹⁰.

Apert syndrome is characterized by bicoronal synostosis, midface hypoplasia, and complex syndactyly involving both hands and feet (Table 1). The abnormal skull morphology is a result of maldevelopment of brain calvarias and cranial base ⁶⁴. This fusion of the digits differentiates Apert syndrome from other FGFR2-related syndromes. Apert syndrome mutations have been shown to follow autosomal dominant inheritance ^3,65,66. Apert syndrome is typically caused by specific missense mutations of the highly-conserved linker region between IgII and IgIII domains of FGFR2, most often either Ser252Trp or Pro253Arg ^16,43,67-69. Cleft palate can also occur in this disorder, most commonly in those with Ser252Trp mutations ⁴³. Other studies corroborate the findings of Ser252Trp and Pro253Arg mutations in majority of patients with FGFR2 related Apert syndrome, although Ser252Trp variant is more common, and Ser252Phe mutations have also been reported ^44,70-72. Interestingly, unlike the Ser252Trp variant, the Pro253Arg mutation results in generalized increased affinity of FGFs to FGFR2 ^28,73. Such receptor mutations exhibit selective decrease in the dissociation kinetics of FGF2 but not of other FGF ligands examined ⁷⁴. Recently, Andreou and colleagues reported a young girl with Pro253Arg FGFR2 mutation-associated Apert syndrome who also had early-onset low-grade papillary carcinoma ⁷⁵. Studies by Lomri et al. on Apert syndrome provided evidence to suggest that FGFR2 mutations resulted in greater numbers of precursor cells entering the osteogenic pathway, which lead to premature calvarial ossification and bone matrix formation ⁷⁶. FGFR2 mutations may result in premature suture fusion through upregulation of epidermal growth factor (EGF) and platelet derived growth factor (PDGF) α signaling ⁷⁷.

Antley-Bixler is a rare syndrome associated with multisuture synostosis as well as a broad spectrum of other congenital anomalies including midface hypoplasia, radiohumeral synostosis, joint contractures and arachnodactyly (Table 1) ^78-80. Prognosis is poor and mortality in the newborn is high, largely from airway issues. Although the syndrome has been described for several decades now, the molecular basis is evolving ⁸¹. Antley-Bixler syndrome was thought to have autosomal recessive inheritance, but more recently evidence has emerged of a dominant de novo mutation in FGFR2 resulting in a Ser351Cys amino acid substitution in the IgIII domain ^82,83. These findings suggest that there may be both an autosomal recessive and dominant form.

Beare-Stevenson cutis gyrata syndrome is characterized by several anomalies in addition to the skin condition of redundant and convoluted folds, including craniosynostosis, ear defects, anogenital anomalies (Table 1) ^84-86. The syndrome follows an autosomal dominant inheritance pattern. Although the syndrome is exceedingly rare with only a few cases reported, the majority of the patients with craniosynostosis had severe Kleeblattschadel deformity ^87-90. Beare-Stevenson is caused by de novo heterozygous mutations in FGFR2, including Tyr375Cys mutation in the transmembrane domain, and Ser372Cys mutation in the carboxy-terminal end of the linker region between IgIII and transmembrane domains ^90-94. These activating point mutations account for the majority of Beare-Stevenson syndrome cases. A de novo heterozygous 63-bp deletion in the FGFR2 was also noted in a boy presenting with Beare-Stevenson, and the authors proposed that such a deletion would alter the splicing of isoform IIIc ⁹⁵. However, in the same report, no mutations were found in a few patients with the syndrome, suggesting genetic heterogeneity.

Jackson-Weiss syndrome is another rare craniosynostostic disorder, with variable phenotypes, inherited in an autosomal dominant fashion (Table 1). Patients with this syndrome also demonstrate facial dysmorphism and foot malformations including tarsal/metatarsal fusion, although the hands are normal in appearance ⁹⁶. Further investigation of one of the families described in the original report of Amish kindreds led to the identification of Ala344Gly mutation in the conserved region of IgIII domain of FGFR2 ⁹⁷. Similarly, other reports have confirmed missense mutations in FGFR2 exons IIIa and IIIc in Jackson-Weiss, including a Glu289Pro mutation in exon IIIa ^98,99. A novel FGFR2 Ser347Cys mutation was identified in a Japanese patient with Jackson-Weiss syndrome ¹⁰.

De novo FGFR2 mutations have been identified in Bent Bone Dysplasia (BBDS), a rare occurring perinatal lethal skeletal dysplasia of autosomal dominant inheritance. BBDS is characterized by craniosynostosis, dysmorphic oral maxillofacial features as well as hypoplastic pubis, clavicles, osteopenia, and bent long bones ^100,101. One study demonstrated that four individuals affected with this disorder were heterozygous for missense mutations that resulted in the addition of a polar amino acid to the hydrophobic transmembrane region of the receptor ¹⁰⁰. In BBDS, unique mutations reduce receptor levels at the plasma membrane and decrease responsiveness to FGF2, therefore attenuating canonical signaling. Heterozygosity for the de novo missense mutations Tyr381Asp and Met391Arg has been reported ¹⁰⁰. More recently, in addition to canonical signaling at the plasma membrane, FGFR2 has been shown to regulate bone formation within the nucleolus ⁴⁰. These aforementioned mutations also enhance nucleolar occupancy of FGFR2 at the ribosomal DNA (rDNA) promoter ⁴⁰.

Seathre-Chotzen is yet another autosomal dominant craniosynostotic condition most commonly associated with mutations in the TWIST gene (Table 1). However, Paznekas and colleagues reported on a single case with a val-val deletion in FGFR2 ¹⁰². Several years later, a Gln289Pro mutation in FGFR2 was detected in a family with Saethre-Chotzen syndrome ¹⁰³. The spectrum of craniosynostosis observed may be either uni- or bicoronal, with sagittal suture fusion exceedingly rare and metopic suture fusion found in few reports ¹⁰⁴. Other features of the disorder include short stature, including syndactyly of hands and feet, hypertelorism, ptosis, and abnormalities of the spine.

An unspecified craniosynostosis syndrome, craniofacial-skeletal-dermatological dysplasia, has been associated with a Trp290Cys mutation in the extracellular domain of FGFR2 ¹⁰⁵. This syndrome is characterized by complex multiple suture craniosynostosis, hydrocephalus, corneal scleralization, atresia of auditory meatus, choanal stenosis, broad thumbs and great toes, scoliosis, acanthosis nigricans, and mental retardation.

Syndrome management

Management of syndromic craniosynostosis is optimized by a multidisciplinary approach involving plastic and reconstructive surgeons, neurosurgeons, neurologists, pediatricians, and genetic counselors. Various imaging modalities are used to better facilitate prenatal diagnosis including ultrasound, magnetic resonance imaging (MRI) and genetic testing ¹⁰⁶. Ultrasound may be used to look for associated limb anomalies and craniofacial dysmorphologies that may point to a particular syndrome. 3-D ultrasound and MRI should be used to further assess findings suspicious for a craniosynostosis syndrome on two-dimensional ultrasound ^53,107. MRI is particularly helpful in evaluating associated central nervous system anomalies.

Given the autosomal dominant inheritance pattern of FGFR2-related craniosynostosis syndromes, genetic counseling is critical once the diagnosis is confirmed. Follow-up should be scheduled regularly to monitor for signs and symptoms of increased intracranial pressure. Once a pathogenic variant has been identified, prenatal testing is possible, although the predictive value is poor. Non-invasive means such as the use of cell-free fetal DNA present in the maternal blood to assess for known pathogenic gene mutations is possible in the first trimester ¹⁰⁸. Amniocentesis and chorionic villus sampling are invasive procedures traditionally use for prenatal screening. Testing for pathogenic variants of genes such as FGFR2 is indicated in high-risk pregnancies with family histories significant for craniosynostosis, but routine fetal ultrasound is sufficient for low-risk pregnancies ¹⁰⁶.

Treatment is typically geared towards surgical correction of the clinical manifestations, the prevention of future deformation, while reducing the risk of increased intracranial pressure. Surgery is done emergently in cases of increased intracranial pressure. Otherwise, the optimal timing for fixation remains controversial, with some surgeons preferring early intervention to allow for expansion of a constricted brain and corneal protection, and others delaying surgery to allow for more stability of the skull in order to decrease the need for revisional surgery ¹⁰⁶. The goal of surgery is to improve skull, orbit, and facial deformities. Whereas experience has shown that single suture craniosynostosis can often be managed successfully with a single surgical procedure, complex/multiple suture craniosynostosis often necessitates a staged approach ¹⁷. Surgical correction methods include open calvarial reconstruction, distraction, and less invasive techniques such as regional cranioectomy, liberation of occipital bone flap and spring-mediated cranial vault remodeling ^106,109. Increasing experience at the Children's Hospital of Philadelphia, a high-volume center for syndromic craniosynostosis surgery, has favored posterior cranial vault distraction over fronto-orbital advancement as the initial cranial remodeling procedure ¹⁰⁶. Posterior cranial vault distraction has several advantages including superior cranial volume expansion, improvements in anterior and posterior cranial morphology, with favorable morbidity profile ¹⁰⁶. Early data in patients with Apert syndrome suggests that the overall number of major craniofacial procedures within the first 5 years of life is decreased following early posterior cranial vault distraction ¹⁰⁶.

Pharmacologic blockade of craniosynostotic pathways

There is a developing interest in inhibiting the molecular pathways involved in the development of craniosynostosis in the aforementioned syndromes. A soluble form of FGFR2 with Ser252Trp mutation inhibits osteoblastic differentiation caused by gain-of-function mutations in FGFR2 in an Apert mouse model, and partially prevents craniosynostosis ^110-112. Further attenuation of the signaling pathways has also been shown to prevent premature suture fusion ¹¹³. Uncoupling of the docking protein FRS2 and activated FGFR2 (Figure 1) leads to normal skull development in a murine model of Crouzon-like craniosynostosis ¹¹³. Using a murine model, Xu and colleagues showed that an activating mutation in FGFR2 results in increased trabecular bone formation and decreased bone resorption through a Wnt/β-catenin-dependent anabolic effect ¹¹⁴. Pharmacologic inhibition of Wnt/β-catenin signaling partially reverses this process ¹¹⁴. These early studies provide support for developing agents that inhibit the pathways that lead to craniosynostosis. Also, knowing that patients previously diagnosed with FGFR2-related Crouzon syndrome had ERF-related craniosynostosis, ERF may serve as an target for therapeutic modulation ^32,115. In a mouse model of craniosynostosis, Shukla and colleagues demonstrated that a small hairpin RNA targeting the dominant mutant form of FGFR2 prevents craniosynostosis in mice ^116,117. Pharmacologic blockade of mitogen-activated protein (MAP) kinase 1 and 2 (MEK1/2)/ERK pathway (Figure 1) by U0126 in mutant mice significantly inhibits craniosynostosis ¹¹⁶. These authors proposed that small-molecules could potentially be used for both prevention and therapy.

Compared to the investigations underway for FGFR-related malignancies, less progress has been made with regard to therapeutic strategies for FGFR-related skeletal and craniosynostotic disorders. Tyrosine kinase inhibition is one such promising strategy in mutated FGFR-related craniosynostotic conditions ¹¹⁸. PD173074, an FGFR tyrosine kinase inhibitor, has been used in a model of FGFR2 related Crouzon syndrome, demonstrating coronal suture patency and no evidence of craniosynostosis ¹¹⁹. Wang et al. observed ligand-independent phosphorylation of FGFR2 in a model of Beare-Stevenson cutis gyrate syndrome and p38 MAPK pathway activation ¹²⁰. The use of p38 kinase inhibitor suppressed premature suture closure in mutated mice ¹²⁰. Recently, a tyrosine kinase inhibitor (Figure 1), ARQ 087, has been shown to rescue the major pathological effects of activating mutations in FGFR1, FGFR2, and FGFR3 in experimental models of chondrodysplasias and craniosynostosis ¹²¹.

Conclusions and future aspects

The FGFR2 is perhaps the most extensively studied gene that is mutated in various craniosynostotic syndromes including Crouzon, Apert, Pfeiffer, Antley-Bixler, Beare-Stevenson cutis gyrata, Jackson-Weiss, Bent Bone Dysplasia, and Seathre-Chotzen-like syndromes. The majority of these mutations are missense mutations that result in constitutive activation of the receptor and downstream molecular pathways. Treatment involves a multidisciplinary approach with ultimate surgical fixation of the cranial deformity to prevent further sequelae. Understanding the molecular mechanisms has allowed for the investigation of different therapeutic agents that can potentially be used to prevent the disorders. Genetic studies are recommended for not only individuals with known syndromic features, but also for family members with uncommon or normal phenotypes to improve counseling outreach.

One of the more fascinating questions in the biology of the FGFR2 gene is why mutations in this single gene can give rise to such a wide range of disorders with such a broad spectrum of phenotypes. As seen in the preceding sections, some disorders involve single suture synostosis, whereas others affect multiple sutures. Some disorders are principally confined to the craniofacial skeleton, whereas others have prominent extracranial dysmorphologies. Given the extensive study of FGFR2 mutations in clinical samples, this gene could prove an interesting test case for systematically mapping genotype-phenotype relationships in a highly conserved pathway. It will be interesting to discover whether various mutations in FGFR2 and its pathway members have subtly different activities, and how these activities give rise to the observed spectrum of phenotypes. Another possibility is that modifier genes alter the phenotypic response in patients to a common array of gain or loss of function mutations. The accumulation of patient sequencing data and more studies in animal models will be important to sorting out these possibilities.

Further research efforts are need to better understand screening and effective methods of early intervention and prevention. Also, knowing that some individuals with mutations may be phenotypically normal and have syndromic offspring, a better understanding of the genetic, epigenetic, and environment interplay is critical. When deciding surgical options for patients with craniosynostosis, more data is necessary to guide optimal timing of intervention as well as the best method of cranial remodeling to allow for minimal morbidity and less need for future interventions.