American Journal of Medical Genetics Part C (Seminars in Medical Genetics) 163C:259–270 (2013) A R T I C L E Genotype and Clinical Care Correlations in Craniosynostosis: Findings From a Cohort of 630 Australian and New Zealand Patients T. ROSCIOLI,* G. ELAKIS, T.C. COX, D.J. MOON, H. VENSELAAR, A.M. TURNER, T. LE, E. HACKETT, E. HAAN, A. COLLEY, D. MOWAT, L. WORGAN, E.P. KIRK, R. SACHDEV, E. THOMPSON, M. GABBETT, J. MCGAUGHRAN, K. GIBSON, M. GATTAS, M‐L. FRECKMANN, J. DIXON, L. HOEFSLOOT, M. FIELD, A. HACKETT, B. KAMIEN, M. EDWARDS, L.C. ADÈS, F.A. COLLINS, M.J. WILSON, R. SAVARIRAYAN, T.Y. TAN, D.J. AMOR, G. MCGILLIVRAY, S.M. WHITE, I.A. GLASS, D.J. DAVID, P.J. ANDERSON, M. GIANOUTSOS, AND M.F. BUCKLEY Craniosynostosis is one of the most common craniofacial disorders encountered in clinical genetics practice, with an overall incidence of 1 in 2,500. Between 30% and 70% of syndromic craniosynostoses are caused by mutations in hotspots in the fibroblast growth factor receptor (FGFR) genes or in the TWIST1 gene with the difference in detection rates likely to be related to different study populations within craniofacial centers. Here we present results from molecular testing of an Australia and New Zealand cohort of 630 individuals with a diagnosis of craniosynostosis. Data were obtained by Sanger sequencing of FGFR1, FGFR2, and FGFR3 hotspot exons and the TWIST1 gene, as well as copy number detection of TWIST1. Of the 630 probands, there were 231 who had one of 80 distinct mutations (36%). Among the 80 mutations, 17 novel sequence variants were detected in three of the four genes screened. In addition to the proband cohort there were 96 individuals who underwent predictive or prenatal testing as part of family studies. Dysmorphic features consistent with the known FGFR1‐3/TWIST1‐ associated syndromes were predictive for mutation detection. We also show a statistically significant association between splice site mutations in FGFR2 and a clinical diagnosis of Pfeiffer syndrome, more severe clinical phenotypes associated with FGFR2 exon 10 versus exon 8 mutations, and more frequent surgical procedures in the presence of a pathogenic mutation. Targeting gene hot spot areas for mutation analysis is a useful strategy to maximize the success of molecular diagnosis for individuals with craniosynostosis. © 2013 Wiley Periodicals, Inc. KEY WORDS: fibroblast growth factor receptor; TWIST1; crouzon; pfeiffer; apert; Saethre–Chotzen; Muenke How to cite this article: Roscioli T, Elakis G, Cox T, Moon D, Venselaar H, Turner A, Le T, Hackett E, Haan E, Colley A, Mowat D, Worgan L, Kirk EP, Sachdev R, Thompson E, Gabbett M, McGaughran J, Gibson K, Gattas M, Freckmann ML, Dixon J, Hoefsloot L, Field M, Hackett A, Kamien B, Edwards M, Adès L, Collins F, Wilson M, Savarirayan R, Tan T, Amor D, McGIllivray G, White S, Glass I, David DJ, Anderson PJ, Gianoutsos M, Buckley MF. 2013. Genotype and clinical care correlations in craniosynostosis: Findings from a cohort of 630 Australian and New Zealand patients. Am J Med Genet Part C Semin Med Genet 163C:259–270. The authors did not have a conflict of interest regarding this study. Tony Roscioli is a clinical geneticist with an interest in craniofacial, immunodeficiency and neuronal migration disorders, and the application of clinical genomics to gene identification and patient care. George Elakis is a senior molecular scientist with 23 years laboratory experience. His interests include the molecular genetics of craniofacial and intellectual disability disorders, next generation sequencing and computational laboratory support networks. Timothy C. Cox is Professor and Laurel Endowed Chair in Craniofacial Research in the University of Washington's Department of Pediatrics (Division of Craniofacial Medicine) and the Center for Developmental Biology and Regenerative Medicine at Seattle Children's Research Institute. He research interests lie in understanding genetic and epigenetic contributions to craniofacial development and dysmorphism. David Moon is a craniofacial surgeon who is working at Sydney Children's Hospital to gain additional clinical experience in plastic surgery. Hanka Venselaar is a structural bioinformatician at the CMBI in Nijmegen, the Netherlands. She focuses on the analysis of mutations and their effects on protein structures. Anne Turner is the director of the Department of Medical Genetics at Sydney Children's hospital. She has an interest in dysmorphology, eye and craniofacial disorders. Trang Le is a molecular scientist working in the South Eastern Area Laboratory Services molecular diagnostic laboratory. Emma Hackett is a molecular scientist working in the South Eastern Area Laboratory Services molecular diagnostic laboratory and a senior member of the next generation sequencing diagnostic service. ß 2013 Wiley Periodicals, Inc. 260 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) ARTICLE Eric Haan is a clinical geneticist in the South Australian Clinical Genetics Service, a unit of SA Pathology based at Women's and Children's Hospital, Adelaide and a Clinical Affiliate Professor in the Department of Paediatrics, University of Adelaide. Academic interests include the genetics of intellectual disability and cerebral palsy, and the causes and prevention of genetic disorders and birth defects. Alison Colley is the director of the Department of Medical Genetics at Liverpool hospital and she has expertise in dysmorphology. David Mowat is a clinical geneticist at Sydney Children's Hospital with an interest in syndrome diagnosis. Lisa Worgan is a clinical geneticist at the Department of Medical Genetics at Liverpool hospital and she has expertise in dysmorphology. Edwin Kirk is a clinical geneticist working at Sydney Children's Hospital. His research interests include the genetics of congenital heart disease and of facial clefting. Rani Sachdev is a clinical geneticist working at Sydney Children's Hospital. Her interests include the genetics of epileptic encephalopathies and clinical dysmorphology. Elizabeth Thompson is a Clinical geneticist with the South Australian Clinical Genetics Service based at the Women's and Children's Hospital. Her main research interests include dysmorphology and skeletal dysplasias. She is an Associate Professor with the Department of Paediatrics in the University of Adelaide. Michael Gabbett is a senior staff specialist in clinical genetics at the Royal Brisbane & Women's Hospital, Australia. Julie McGaughran is a clinical geneticist with an interest in dysmorphology and cardiac genetics. She is the director of Genetic health Queensland and a member of the council of ESHG. Kate Gibson is a Clinical Geneticist working in the South Island Hub of Genetic Health Service NZ. Michael Gattas is a Clinical Geneticist with Genetic Health Queensland, and in private practice in Brisbane, Australia. Mary‐Louise Freckmann is a clinical geneticist who has an interest in clinical dysmorphology. Joanne Dixon is the National Clinical Director, Genetic Health Service NZ. She has an interest in the provision and quality assessment of clinical genetic services. Lies Hoefsloot is a clinical molecular geneticist working at the Nijmegen molecular diagnostic laboratory. Her expertise includes the provision of molecular and next generation sequencing diagnostic services. Michael Field is a clinical geneticist working primarily in the area of X linked intellectual disability (XLID) as part of the GOLD service. The service has had considerable success with the diagnosis and management of families with XLID. Anna Hackett is a clinical geneticist working primarily in the area of X linked intellectual disability (XLID) as director of the GOLD service. Benjamin Kamien is a clinical geneticist at Hunter Genetics. His clinical and research areas of interest include dysmorphology, Opitz syndrome, and overgrowth syndromes. Matthew Edwards is a clinical geneticist with a long‐standing interest in dysmorphology. Lesley Ades is a clinical geneticist at the Children's hospital at Westmead with an interest in connective tissue disorders. Felicity Collins is Head of Department of Clinical Geneticis at the Children's hospital at Westmead. Her interests are in clinical dysmorphology, connective tissue disorders, skeletal dysplasias and prenatal diagnosis. Meredith Wilson is a clinical geneticist based at Children's Hospital Westmead, Sydney. Dr. Wilson has more than 25 years' experience in clinical genetics practice, with an emphasis on clinical dysmorphology. Dr. Wilson has contributed to a number of publications regarding clinical and molecular aspects of syndrome diagnosis. Ravi Savarirayan is a clinical geneticist with expertise in skeletal dysplasias and a research interest in the targeted management of genetic disorders. Tiong Yang Tan is a clinical geneticist with research interest in the developmental mechanisms underpinning dysmorphology and genetic syndromes. David Amor is a clinical geneticist and Director of Victorian Clinical Genetics Services. His special interests include chromosome biology, the identification of genes for rare disorders, and genetic factors associated with assisted reproduction. George McGillivray is a clinical geneticist working in clinical practice in Melbourne Australia. He previously consulted in the Craniofacial Unit at the Royal Children's Hospital. He currently consults at the perinatal clinics at the Royal Women's Hospital and the Mercy Hospital for Women with a focus on prenatal diagnosis and in the Neurogenetics Clinic at the Royal Children's Hospital. Susan White is a clinical geneticist in Melbourne, Australia with a research interest in dysmorphic syndromes. She assists in the curation of the POSSUMweb dysmorphology database. Ian Glass is a medical geneticist at the University of Washington, Seattle with a long‐standing interest in understanding genetic disorders of the skeleton and cranium and contributed to this work thorough his efforts in phenotyping, genetic mapping and gene identification. David John David is Clinical Professor of Cranio Maxillofacial Surgery, University of Adelaide, Clinical Professor Macquarie University and Head of Australian Cranio Facial Unit, WCH. Peter Anderson is Associate Professor at the University of Adelaide, and Director of Research at the Australian Craniofacial Unit. He is currently the President of the Asia Pacific Craniofacial Association and is an Active member of the International Society of Craniofacial surgeons. Mark Gianoutsos is the Head of Department of the Sydney Craniofacial Unit at the Sydney Children's Hospital and the Plastic and Craniofacial Research Unit, University of New South Wales. He trained in Craniofacial Surgery at New York University and Is a Founding Member of the Australian and New Zealand Society of Cranio Maxillo Facial Surgeons and a Member of the International Society of Craniofacial Surgeons. Michael Buckley is the director of the SEALS molecular and cytogenetics laboratory. He has an interest in craniofacial diseases, gene identification and the provision of next generation sequencing for clinical services. Grant sponsor: The Australian Cranio‐Maxillo Facial Foundation. *Correspondence to: Dr. T. Roscioli, School of Women's and Children's Health, University of New South Wales, Sydney, NSW 2013, Australia. E‐mail: tony.roscioli@sesiahs.health.nsw.gov.au DOI: 10.1002/ajmg.c.31378 Article first published online in Wiley Online Library (wileyonlinelibrary.com): 11 October 2013 ARTICLE INTRODUCTION Craniosynostosis is a significant cause of childhood morbidity [Wilkie, 1997]. The patency of cranial sutures in childhood is critical for craniofacial development and the accommodation of central nervous system growth. Sutures have a number of important functions including the provision of skull flexibility to allow changes in skull volume for the accommodation of brain growth in early life, the maintenance of rigid connections between adjacent bones, and alignment and fusion of adjacent bones once central nervous system growth is complete. Premature Sutures have a number of important functions including the provision of skull flexibility to allow changes in skull volume for the accommodation of brain growth in early life, the maintenance of rigid connections between adjacent bones, and alignment and fusion of adjacent bones once central nervous system growth is complete. closure of cranial sutures therefore results in marked cranial and/or facial dysmorphism. If untreated, craniosynostosis may lead to major malformation syndromes characterized by alterations in skull morphology, orbital and facial hypoplasia, raised intracranial pressure, deafness, blindness and intellectual disability. Surgical intervention is required to alleviate raised intracranial pressure, and correct calvarial, facial and dental deformation. In the pre‐molecular diagnostic era, the craniofacial and appendicular skeletal features of craniosynostosis syndromes were utilized for clinical AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) classification and diagnosis [Gorlin et al., 1990]. In many cases of FGFR‐ associated craniosynostosis, the presence of skeletal changes provides a clinical means of syndromic assignment [Gorlin et al., 1990]. Among the best characterized of these disorders are Crouzon, Pfeiffer, Apert, Muenke, and Saethre– Chotzen syndromes, which share phenotypic features of premature synostosis of the cranial sutures, underdevelopment of the mid‐face, and variable involvement of the extremities. Currently, surgical management of craniosynostosis is individualized according to the severity and age of onset of craniofacial anomalies, without routine reference to 3D photography or molecular diagnostics. The molecular diagnosis of craniosynostosis syndromes became feasible following the discovery that mutations in the genes encoding FGFR1 and FGFR2 are associated with Crouzon, Pfeiffer, Apert and Jackson‐Weiss syndromes [Jabs et al., 1994; Muenke et al., 1994; Reardon et al., 1994; Rutland et al., 1995; Wilkie et al., 1995; Roscioli et al., 2000]. Rarely, mutations The molecular diagnosis of craniosynostosis syndromes became feasible following the discovery that mutations in the genes encoding FGFR1 and FGFR2 are associated with Crouzon, Pfeiffer, Apert and Jackson‐Weiss syndromes are also identified in MSX2 (variable Boston‐type craniosynostosis) [Jabs et al., 1993]. Other syndromes presenting with craniosynostosis, hypertelorism and nasal anomalies include frontonasal dysplasia/frontorhiny, which may occur secondary to homozygous mutations in ALX3 [Twigg et al., 2009], and craniofrontonasal dysplasia, an X‐linked disorder with paradoxical severity in females due to tissue boundary disruption 261 secondary to mutations in EFNB1 [Twigg et al., 2004]. Additional recognizable syndromes such as Beare– Stevenson, Crouzon with acanthosis nigricans, and Antley‐Bixler (caused by mutations of FGFR2 and POR) have also been described [Beare et al., 1969; Stevenson et al., 1978; Meyers et al., 1995]. In contrast, the genetic etiologies of the nonsyndromic uni‐ or oligosutural craniosynostoses are poorly understood and mutations are identified less frequently. Of these, the most important are the FGFR3 P250R mutation in coronal craniosynostosis and Muenke syndrome [Glass et al., 1994; Muenke et al., 1994; Moloney et al., 1997; Graham et al., 1998] and high frequency mutations described recently in TCF12 and ERF [Sharma et al., 2013; Twigg et al., 2013]. Copy number variants involving chromosome 9p and point mutations in FREM1 have been reported in a low percentage of individuals with metopic craniosynostosis [Vissers et al., 2011] and a recent genome‐wide association study identified variants in linkage disequilibrium with BBS9 and BMP2 in sagittal craniosynostosis [Justice et al., 2012]. Another rare syndrome with primary involvement of the sagittal suture (Sensenbrenner syndrome) is due to de novo WDR35 mutations [Gilissen et al., 2010]. Documentation of the wide spectrum of skeletal anomalies present in the FGFR3‐related Muenke coronal craniosynostosis syndrome (CCS) further underscores a significant role for FGFRs in both endochondral and membranous bony development [Bellus et al., 1996; Moloney et al., 1997; Muenke et al., 1997; Graham et al., 1998]. There is a significant clustering of mutations in some exons of FGFR‐1, ‐2, and ‐3 in contrast to TWIST1 mutations, which occur throughout the coding region. Additional mutations in non‐hotspot areas are identifiable with the analysis of a further 6 exonic regions in FGFR2 and dosage studies for TWIST1 hemizygosity. This complete screen detects all reported mutations including the 80 mutations identified in this report. 262 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) Many single syndrome or case reports with small numbers have been published; however, large cohorts with documented clinical phenotypes and extensive molecular screening, such as are presented here, are uncommon [Ingersoll et al., 2001; Kan et al., 2001; Passos‐Bueno et al., 2008; Wilkie et al., 2010]. Here we describe the mutation spectrum for FGFR1‐3 and TWIST1 in a large cohort of patients with syndromic and nonsyndromic craniosynostosis, present a series of undescribed mutations, and draw conclusions regarding genotype–phenotype correlations and surgical outcomes. MATERIALS AND METHODS Clinical Ascertainment and Statistical Analysis Individuals with craniosynostosis were ascertained through clinical genetics and surgical referrals from five major Australian craniofacial clinics (New South Wales 2, South Australia 1, Queensland 1, Victoria 1) and New Zealand for molecular testing at the SEALS Genetics laboratory, South Eastern Area Laboratory Services, Sydney, Australia. Historical, perinatal, developmental, growth, morphometric and examination data were obtained and collated prior to surgery. In the majority a clinical diagnosis of the type of sutural synostosis and presence or absence of a syndrome was made by a clinical geneticist. The main clinical differentiators for the more common syndromes, other than recognizable facial dysmorphology, included lack of limb features (Crouzon syndrome), broad/deviated thumbs/halluces (Pfeiffer syndrome), brachydactyly/ clinodactyly (Muenke syndrome) and syndactyly (Apert syndrome). The clinical designation of the patient cohort was based on referral information and clinical review, where possible. Where initial referral diagnosis was discordant to that determined by a clinical geneticist, the clinical geneticist classification was utilized. The number and type of surgical procedures were tabulated for each patient according to the year in which they were performed to facilitate inter‐ group analyses. Statistical difference in mutation detection, within and between syndromic and non‐syndromic groups and surgical classes, were calculated using an online Chi squared calculator (www.vassarstats.net). DNA Analysis Mutation screening was performed by Sanger sequencing analysis of FGFR1, exon 7, FGFR2 exons 3, 5, 8, 10, 11, and 14–17, FGFR3 exons 7 and 10 and TWIST1 [nomenclature based on Ingersoll et al., 2001; Kan et al., 2001]. Patient DNA was prepared from peripheral blood collected in EDTA using a Qiagen DNA extraction kit, following the manufacturer’s instructions and stored at 4°C at a concentration of 100 ng/ml until used. Patient DNA was amplified using the polymerase chain reaction (PCR) under routine conditions. DNA sequencing was performed using standard Sanger sequencing protocols and the products analyzed by capillary electrophoresis on an ABI3130xl Genetic Analyser. All analyses included amplicon‐specific positive and negative controls. Oligonucleotide primers (available on request) were designed for the above exons based on the FGFR1 (NM_ 023110.2, ENSP00000400162), FGFR2 (NM_000141.4, ENSP00000351276), FGFR3 (NM_000142.4, ENSP00000414914) and TWIST1 (NM_000474.3, ENSP00000242261) reference sequences. Structural Analysis of Novel Mutations The effects of the mutations were studied in detail using the partly solved protein structures of FGFR2 and FGFR3, and a homology model for TWIST1. The structures were analysed using the YASARA and WHAT IF Twinset [Vriend, 1990; Krieger et al., 2002]. The FGFR2 mutations were studied using PDB‐file 1ev2. This file contains the extracellular ligand binding domain of FGFR2 in combination with FGF2. Almost all mutations were located in this domain, except for K367E. The FGFR3 mutation P260L was ARTICLE analysed using PDB‐file 1ry7. This file contains the extracellular ligand binding domain of FGFR3 in complex with FGF3. The other mutation in FGFR3, R397C, was not present in the structure. The TWIST1 protein does not have a solved X‐ray structure, therefore a homology model was built using the helix‐loop‐helix motif in Myc proto‐ oncogene as a template (PDB‐file 1nkp). The model was built using the YASARA modeling script with standard parameters [Krieger et al., 2002]. The model consists of two monomeric helix‐loop‐ helix chains (residues 94–198) that bind a DNA strand. RESULTS Mutation Findings In the period 2001–2013, a total of 630 probands referred for molecular diagnosis had mutation tests targeted to the “hot‐spot” regions of the FGFR1, FGFR2, FGFR3 genes and full sequencing of TWIST1. Three hundred patients (231 probands) were identified to have one of a total of 80 different mutations in FGFR2, FGFR3, and TWIST1, including 17 novel sequence variants (Table I), with an overall mutation detection frequency of 36%. Phenotypes in Relation to Mutation Analysis Results The mutation frequencies varied by clinical phenotype with the highest being reported in Apert (100%), Pfeiffer (79.5%), Crouzon (75.6%), syndromic coronal craniosynostosis/Muenke (68.6%) and in Saethre–Chotzen (63.7%) (Table II). These findings confirm the high clinical utility of targeted mutation screening in syndromic craniosynostosis. Novel Mutations Review of the Human Gene Mutation Database, the 1,000 Genomes Project, the University of Washington exome variant server and the published literature led us to categorize 17 variants identified in this cohort as novel. Of these, eight were missense mutations, six ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) 263 TABLE I. Mutations and Clinical Phenotypes in 300 Craniosynostosis Patient Referrals 2001–2013 Gene and mutation FGFR1 Clinical spectrum FGFR1 c.755C>G p.Pro252Arg 2 Pfeiffer, 2 Muenke FGFR1 c.899T>C p.Ile300Thr 2 plagiocephaly, FGFR2 c.314A>G p.Tyr105Cys Gene and mutation § Clinical spectrum FGFR2 c.1099A>G p.Lys367Glu 2 Crouzon FGFR2 c.1115C>G p.Ser372Cys 1 Beare–Stevenson 4 Crouzon FGFR2 c.1124A>G p.Tyr375Cys 1 Beare–Stevenson FGFR2 c.755C>G p.Ser252Trp 34 Apert FGFR2 c.1150G>A p.Gly384Arg 7 Crouzon FGFR2 c.758C>G p.Pro253Arg 13 Apert § FGFR2 c.758C>T p.Pro253Leu 2 bicoronal § FGFR2 c.796G>C p.Ala266Pro 1 Crouzon FGFR2 c.1694A>C p.Glu565Ala 1 Crouzon, 1 Pfeiffer FGFR2 c.799T>C p.Ser267Pro 2 Crouzon FGFR2 Alu insert 2 Apert FGFR2 c.820_824delinsTT 1 atypical Crouzon/ FGFR3 FGFR3 c.749C>G p.Pro250Arg 49 Muenke # † FGFR3 c.779C>T p.Pro260Leu 1 unicoronal, ?Muenke FGFR3 c.847C>A p.Pro283Thr 1 bicoronal 1 bicoronal FGFR2 † p.Val274_Glu275delinsLeu † FGFR2 c.1139_1196dup p.Met400fs 53 1 sagittal FGFR2 c.1576A>G p.Lys526Glu 1 sagittal/Crouzon, mild Costello features FGFR2 c.826T>G p.Phe276Val 1 Crouzon, 1 Pfeiffer FGFR2 c.833G>T p.Cys278Phe 12 Crouzon, 1 multisuture, (non‐penetrant in father) 1 Pfeiffer FGFR2 c.842A>G p.Tyr281Cys 1 Saethre–Chotzen‐like FGFR3 c.1172C>A p.Ala391Glu 2 Crouzon, 1 Crouzon with acanthosis nigricans, 1 unicoronal FGFR2 c.844A>T p.Ser282Cys † 1 Crouzon, mild FGFR2 c.863T>A p.Ile288Asn 2 Crouzon FGFR2 c.868T>C p.Trp290Arg 3 Crouzon FGFR2 c.869G>C p.Trp290Ser 2 Crouzon FGFR2 c.870G>T p.Trp290Cys 1 Pfeiffer FGFR2 c.870G>C p.Trp290Cys 1 Pfeiffer TWIST1 TWIST gene deletion § † † 2 Saethre–Chotzen Twist Del Exon1 1 Saethre–Chotzen TWIST1 c.1‐6C>G 1 Saethre–Chotzen TWIST1 c.31_32del p.Ser11Alafs 226 1 bicoronal TWIST1 c.79C>T p.Gln27 5 Saethre–Chotzen TWIST1 c.81_82delinsTT 2 Saethre–Chotzen p.Gln27_Ser28ins FGFR2 c.940‐1G>C 1 Pfeiffer TWIST1 c.82C>T p.Gln28 3 Saethre–Chotzen FGFR2 c.940‐1G>A 1 Pfeiffer TWIST1 c.94G>A p.Gly32Ser 1 Saethre–Chotzen, 1 sagittal/metopic, 1 metopic, 5 craniosynostosis unspecified † FGFR2 c.940‐2A>G 5 Pfeiffer TWIST1 c.190_200dup p.Ser68Thrfs 61 FGFR2 c.940‐3T>G 1 Pfeiffer TWIST1 c.230delA p.Lys77Serfs 48 2 Saethre–Chotzen FGFR2 c.943G>T p.Ala315Ser 2 Crouzon TWIST1 c.309C>G p.Tyr103 3 Saethre–Chotzen FGFR2 c.992A>T p.Asn331Ile 1 Pfeiffer TWIST1 c.355C>T p.Gln119 1 Saethre–Chotzen FGFR2 c.1007A>G p.Asp336Gly 1 Crouzon TWIST1 c.362C>T p.Thr121Ile 2 Saethre–Chotzen FGFR2 c.1013G>A p.Gly338Glu 2 Crouzon TWIST1 c.368C>A p.Ser123 2 Saethre–Chotzen FGFR2 c.1018T>C p.Tyr340His 8 Crouzon TWIST1 c.371T>C p.Leu124Pro 1 Saethre–Chotzen FGFR2 c.1018T>A p.Tyr340Asn 4 Crouzon TWIST1 c.376G>T p.Glu126 1 Saethre–Chotzen FGFR2 c.1019A>G p.Tyr340Cys 2 Pfeiffer TWIST1 c.380C>T p.Ala127Val 1 Saethre–Chotzen FGFR2 c.1024T>C p.Cys342Arg 6 Pfeiffer, TWIST1 c.395G>T p.Arg132Leu 1 Saethre–Chotzen † 1 Saethre–Chotzen 1 cloverleaf skull TWIST1 c.397_418dup p.Ser140 4 Saethre–Chotzen 8 Crouzon, 3 Pfeiffer TWIST1 c.405_425dup 5 Saethre–Chotzen FGFR2 c.1026C>G p.Cys342Trp 8 Crouzon TWIST1 c.407C>T Pro136Leu 1 Saethre–Chotzen FGFR2 c.1032G>A p.Ala344Ala 5 Crouzon, mild TWIST1 c.408dup p.Thr137Hisfs 101 2 Saethre–Chotzen FGFR2 c.1037_1060del 1 Pfeiffer TWIST1 c.401_421dup 1 Saethre–Chotzen FGFR2 c.1025G>C p.Cys342Ser 4 Pfeiffer, 1 Crouzon FGFR2 c.1025G>A p.Cys342Tyr † p.Asp141_Lys142ins7 † p.Asn346_Ser354delinsThr FGFR2 c.1040C>G p.Ser347Cys † p.Ser140_Asp141ins7 3 Crouzon TWIST1 c.421G>C p.Asp141His 1 Saethre–Chotzen (Continued) 264 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) ARTICLE Table 1. (Continued) Gene and mutation FGFR2 c.1052C>G p.Ser351Cys Clinical spectrum Gene and mutation † 1 Antley‐Bixler, Clinical spectrum  TWIST1 c.438dup p.Gln147Serfs 91 4 Saethre–Chotzen 1 Beare–Stevenson, 1 Pfeiffer FGFR2 c.1061C>G p.Ser354Cys 3 Crouzon TWIST1 c.445C>T p.Leu149Phe 2 Saethre–Chotzen FGFR2 c.1084þ1G>A 1 Crouzon TWIST1 c.454G>C Ala152Pro 2 Saethre–Chotzen FGFR2 c.1084þ3A>G 1 Crouzon TWIST1 c.490C>T p.Gln164 1 Saethre–Chotzen Unmarked, previously reported mutation; †, novel pathogenic mutation; §, novel variant of uncertain significance; #, novel likely polymorphism. mutations were associated with premature protein truncation (4 frameshift and 2 nonsense), there were two insertion/ deletion mutations, and one located in a non‐coding region. Pathogenicity for the eight missense mutations was assessed used a combination of in silico software programs Condel and Align GVGD [Tavtigian et al., 2008; González‐Pérez and López‐Bigas, 2011]. Condel was selected as it provides a method for the integration of data from Polyphen‐2, SIFT and Mutation Assessor, which generates higher ROC characteristics compared with their individual use. These analyses were supplemented by protein structure predictions where appropriate protein models exist, as the analysis of the effect of amino acid substitution mutations is informative with regard to mutation pathogenicity (Fig. 1). Mutation by Amino Acid Position in FGFR2 Five novel FGFR2 missense variants were identified, all of which were located in the 3rd Ig‐like domain of the protein. p.Pro253Leu (Bi‐Coronal, 2 Patients) This mutation involves the same cytosine nucleotide in codon 253 which, when mutated to a guanosine, is associated with Apert syndrome. The mutation is located in the linker that connects Ig‐like domains II and III. The side‐chain is solvent‐exposed and the residue could be important for the correct shape of the linker and possibly also for its interactions with the FGF ligand. In silico prediction software analysis of the substitution of a leucine at this position is predicted by Condel to be deleterious (P ¼ 0.934), and category C0 by Align GVGD. It is therefore classified as a variant of unknown significance (VOUS). p.Ala266Pro (Crouzon, 1 Patient) This mutation is located in the 3rd Ig‐ like domain of FGFR2. The side‐chain TABLE II. Percentage Mutation Detection in Different Classes of Craniosynostosis by Gene Crouzon Pfeiffer Apert Muenke Saethre–Chotzen Multisuture Beare–Stevenson Antley‐Bixler Sagittal Coronal Metopic Lambdoid Plagiocephaly Other FGFR1 FGFR2 FGFR3 3 (8%) 66 (73%) 27 (69%) 42 (95%) 2 (2%) 1 (3%) Other 2 (5%) 33 (69%) 1 (1%) 1 (2%) 1 (6%) 1 (25%) 1 (17%) 2 (2%) 2 (2%) 39 (62%) 3 (3%) 2 (2%) 143 39 41 1 (17%) 1 (20%) 5 Totals TWIST 231 No mutation Total 22 (75%) 8 (21%) 0 (0%) 15 (31%) 23 (36%) 16 (94%) 3 (75%) 4 (66%) 92 (98%) 94 (94%) 41 (100%) 5 (100%) 4 (80%) 72 (100%) 90 39 44 48 63 17 4 6 94 102 41 5 5 72 399 630 3 ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) 265 Figure 1. Effect of selected missense mutations on protein structure of FGFR2 (A–E) and FGFR3 (F). Wild‐type amino acids shown in green and mutated amino acids in red. The FGFR2 3rd Ig‐like C2 domain shown in gray and FGF2 ligand shown in khaki. The magenta dots where visualized indicate a loop that was not solved in the protein structure (residues 268–272). A: The missense mutation FGFR2 p. Pro253Leu. B: The missense mutation FGFR2 p.Ala266Pro. C: The missense mutation FGFR2 p.Tyr281Cys. D: The missense mutation FGFR2 p.Trp290Ser. E: The missense mutation FGFR2 p.Tyr340Asn. F: The missense mutation FGFR3 p.Pro260Leu. of the native alanine is semi‐exposed. Although the slightly bulkier side‐chain of proline can be accommodated in this position, it introduces a different backbone that changes the conformation of the IgIII‐like domain. In silico analysis of the substitution of a proline at position 266 is predicted by Condel to be deleterious (P ¼ 0.910) and category C0 by Align GVGD. It is therefore classified as a VOUS. It makes many hydrophobic interactions with surrounding residues and is essential for the overall stability of the domain. Mutation of this residue is likely to be detrimental for the core structure of the domain and thus impact its function. In silico analysis of the substitution of a serine at position 290 is predicted by Condel to be deleterious (P ¼ 0.996) and category C65 by Align GVGD. It is therefore classified as a pathogenic mutation. p.Trp290Ser (Crouzon, 3 Patients) p.Tyr340Asn (Crouzon, 4 Patients) Trp290 is one of the most important residues in the core of the Ig‐like domains. Tyr340 is also located in the core of the IgIII‐like domain adjacent to the Trp290 (see above). Tyr340 is similarly important as it makes many hydrophobic and stabilizing interactions with surrounding residues. Introduction of the smaller and more hydrophilic asparagine disturbs these interactions and affects the core structure of the domain. Tyrosine 340 is among the most frequently mutated residues in FGFR2. In silico analysis of the substitution of an asparagine at position 340 is predicted by Condel to be deleterious (P ¼ 0.904) and category C65 by Align GVGD. It is therefore classified as a pathogenic mutation. 266 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) p.Lys367Glu (Crouzon, 2 Patients) Lysine 367 is not present in the solved structures. It is located in the linker between the last Ig‐like domain and the transmembrane region. The effect of this mutation on the protein structure is unclear; however replacement of a positively charged lysine by a negatively charged glutamine is likely to have an effect on protein function. In silico analysis of the nucleotide substitution 1099A>G predicts a novel gain of a 30 splice site at this position; however this has yet to be confirmed by mRNA studies. The substitution of a glutamic acid at position 367 is predicted by Condel to be neutral (P ¼ 0.000) and category C0 by Align GVGD. The variant is therefore classified as a VOUS based on its potential effect on mRNA splicing. Mutation by Amino Acid Position in FGFR3 Two novel missense variants were identified in FGFR3. The FGFR3 structure is not completely solved. Data can be found in PDB‐file 1RY7 (residue 32– 353). Modeling of only one of the two novel FGFR3 mutations was performed, as the second novel mutation is located outside the region included in the solved structure. p.Pro260Leu (Muenke/ Uni‐Coronal, 1 Patient) This residue is located in the 3rd Ig‐like domain. This domain has a classic double beta‐sheet fold with a tryptophan and disulfide bond in the middle that provides stability. The side chain of Pro260 is close to the surface but still buried in the domain. The backbone makes a strong turn at this position, which is facilitated by the proline. In the Uniprot‐database, a potential glycosylation site is annotated for the adjacent residue 262. Mutation of this residue will likely not disturb the folding of the domain and the leucine side chain may still be accommodated after a small re‐ localization of surrounding residues. The sidechain of leucine is also hydrophobic so it could still make stabilizing hydrophobic interactions. The local backbone is however altered, and this may change the position of some surface residues that may be important for interactions with other proteins. In silico analysis of the substitution of a leucine at position 260 is predicted by Condel to be neutral (score 0.439) and category C0 by Align GVGD. It is therefore classified as a likely polymorphism. ARTICLE that do interact with the DNA. Substitution of a proline will disturb the DNA binding because the surrounding residues are displaced and the rigidity of the proline side chain will disturb the helix conformation. In silico analysis of the substitution of a proline at position 124 is predicted by Condel to be deleterious (score 1.00) and category C65 by Align GVGD. It is therefore classified as a pathogenic mutation. In summary, among the 17 novel variants identified in this study, 12 were determined to be likely pathogenic mutations, four were classified as variants of uncertain significance and one as a likely polymorphism. p.Pro283Thr (Bi‐Coronal With Partial Penetrance) No protein model was available for this mutation. In silico analysis of the substitution of a threonine at position 283 is predicted by Condel to be deleterious (score 0.936) and category C35 by Align GVGD. It is therefore classified as a likely pathogenic mutation. Effect of FGFR2 Splice Mutations on Phenotype Teebi et al. [2002] have previously reported an association between clinical severity and mutations involving FGFR2 splice sites. We therefore also analyzed clinical features in relation to the presence of an FGFR2 splice mutation in this cohort. A highly significant association between FGFR2 splice site mutations and a clinical diagnosis of Pfeiffer syndrome was identified (Fisher’s exact P < 0.0001, Table III) consistent with prior data. Mutation by Amino Acid Position in TWIST1 The majority of novel variants that were observed in TWIST1 were nonsense and frameshift mutations with high likelihoods of pathogenicity (Table I). A single novel missense variant was identified. p.Leu124Pro (Saethre–Chotzen, 1 Patient) Surgical Procedures by Syndrome/ Mutation Leucine 124 is located in the middle of a helix‐loop‐helix motif. These motifs are known to bind DNA. They form a long helix, connected via a loop to another helix. Two helices form a dimer that binds DNA. The Leu124 side chain is not in direct contact with the DNA but supports residues close to it Next we wished to determine if surgical management of syndromic craniosynostosis patients showed associations with mutation status. Surgical type and timing data were obtained predominantly from the craniofacial clinic at Sydney Children’s Hospital Randwick and were TABLE III. Comparison of Clinical Phenotype Versus FGFR2 Splice Site/ Non‐Splice Site Mutations Non‐splice site Splice site P  0.0001, Fisher’s exact test. Pfeiffer Non‐Pfeiffer 18 7 132 3 ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) reviewed jointly by a craniofacial surgeon and a clinical geneticist. The data were obtained by a clinical review of patient surgical records with tabulation of the timing of surgery during each year of life and a determination of whether the surgery was transcranial (TC) or non‐transcranial (NTC) (i.e., any surgery other than transcranial surgery such as limb, oral or abdominal). The timing of surgery data were combined into five periods determined by age: <1 year (infancy, 47 TC, 5 NTC); 1–5 years (early childhood, 79 TC, 17 NTC); 6– 10 years (childhood, 16 TC, 1 NTC); 11–15 years (late childhood, 10 TC, 0 NTC) and >15 years (young adult, 56 TC, 2 NTC). A total of 233 operative procedures including 208 transcranial and 25 non‐transcranial were identified. A borderline significant difference was identified between the rates of transcranial versus non‐transcranial operations for the syndromic craniosynostosis, with an increased likelihood of limb/ hand procedures in the early childhood years (Table IVa; P ¼ 0.0443). The number of non‐transcranial operations (N ¼ 25) was too small for further statistical analysis (see Discussion Section). Highly significant differences in the timing of surgical procedures were identified among the syndromic craniosynostosis patients who underwent transcranial surgical procedures (Table IVb; P < 0.0001). Analysis of the standardized residuals demonstrates that the majority of the differences can be attributed to comparatively fewer transcranial operations in infancy for Crouzon syndrome (SR ¼ 2.19) compared with Pfeiffer (SR ¼ þ2.05) and Saethre–Chotzen syndromes (SR ¼ þ3.15). There was a trend for an increased number of procedures in Apert syndrome in the childhood years, peaking at age 5–10 years (SR ¼ þ1.87), and with many fewer procedures for Apert syndrome in young adulthood (SR ¼ 267 2.72) compared with Crouzon syndrome (SR ¼ þ1.81). There were no significant differences in the type or timing of surgical procedures based on mutation subclassification in Apert syndrome (P ¼ 0.73) or Crouzon syndrome (P ¼ 0.13). DISCUSSION The molecular genetic reference testing program for craniosynostosis syndromes at South Eastern Area Laboratory Services, Sydney, Australia was established in 2001. The approach used has been to target 11 amplicons containing the vast majority of pathogenic mutations within the four genes FGFR1, 2, 3, and TWIST1 that have been associated with Pfeiffer, Crouzon, Muenke, Apert, Antley‐Bixler, and Saethre–Chotzen syndromes. The outcome of the testing program has been the provision of a comprehensive pre‐ and post‐natal service for the diagnosis of the common TABLE IVA. Comparison Surgical Procedures Timing in the Syndromic Craniosynostoses Procedure 2 Transcranial (x SR) Non‐transcranial (x2 SR) Totals Age <1 year infancy Age 1–5 years early childhood Age 6–10 years childhood Age 11–15 years adolescence Age >15 years young adult Totals 47 (0.007 þ0.09) 5 (00.06 0.25) 79 (0.524, 0.72) 17 (4.358, þ2.09) 16 (0.045, þ0.21) 1 (0.372, 0.61) 10 (0.129, þ0.36) 0 (1.073, 1.04) 56 (0.344, þ0.59) 2 (2.866, 1.69) 208 25 52 96 17 10 58 233 Significance: x2df 4 ¼ 9:78, P ¼ 0.0443. TABLE IVB. Comparison of the Timing of Transcranial Surgical Procedures in the Syndromic Craniosynostoses Clinical phenotype 2 Crouzon (N) (x SR) Pfeiffer (x2 SR) Apert (x2 SR) Saethre–Chotzen (x2 SR) Totals 2 Age <1 year infancy Age 1–5 years early childhood Age 6–10 years childhood 12 (4.807 2.19) 17 (4.198 þ2.05) 6 (1.150 1.07) 12 (9.938 þ3.18) 31 (0.034, þ0.19) 10 (1.110, 1.05) 18 (2.509, þ1.58) 4 (1.065, 1.03) 14 (0.099, 0.32) 6 (0.164, 0.40) 11 (3.491, þ1.87) 1 (1.680, 1.30) 47 63 32 Age 11–15 years adolescence 6 0 4 0 N, number of cases; x , Chi Squared; df, degrees of freedom; SR, standardized residuals. Significance: x2df 12 ¼ 46:78, P ¼ 0.0001. (0.323, þ0.57) (2.212, 1.49) (2.088, þ1.45) (1.058, 1.03) 10 Age >15 years young adult 36 (3.277, 13 (0.031, 2 (7.401, 5 (0.144, 56 þ1.81) þ0.17) 2.72) 0.38) Total 99 46 41 22 208 268 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) craniosynostosis syndromes for families in Australasia, with an increasing uptake of prenatal testing rising from 1–2 to 4–6 cases per year by 2013 consistent with the growth of the program. A total of 630 proband samples (with syndromic and unisutural craniosynostosis) have been studied to date with an overall mutation detection frequency of 36%. Mutation detection frequencies varied according to clinical diagnosis with the highest detection rates in Apert (100%) and Pfeiffer syndromes (79.5%) (Table II). Multi‐suture A total of 630 proband samples (with syndromic and unisutural craniosynostosis) have been studied to date with an overall mutation detection frequency of 36 %. Mutation detection frequencies varied according to clinical diagnosis with the highest detection rates in Apert (100 %) and Pfeiffer syndromes (79.5 %). craniosynostosis without specific dysmorphic features was found to have a low frequency of mutation detection, with one out of 17 referrals having the FGFR2 p.Cys278Phe mutation usually associated with Crouzon syndrome. No other mutations were detected in the multi‐suture cohort. In contrast, a low frequency of mutations was identified with this screen in non‐syndromic unisutural craniosynostosis, other than coronal craniosynostosis, in which a number of mutations, including the FGFR1 c.899T>C p.Ile300Thr (2 plagiocephaly, 1 coronal), FGFR2 p.Pro253Leu (novel, 2 bicoronal), and FGFR3 p.Pro260Leu (novel, 1 unicoronal/Muenke), p.Pro283Thr (novel, reduced penetrance), p.Ala391Glu (2 Crouzon, 1 Crouzon with acanthosis nigricans, 1 unicoronal),a known TWIST1 nonsense mutation, c.368C>A, p.Ser123 (1 Saethre–Chotzen, 1 coronal) and a novel TWIST1 nonsense mutation, c.31_32del p.Ser11Alafs 226, were detected. Some of the patients with coronal craniosynostosis had features suggestive of a mild syndromic presentation; however given phenotypic variability, we recommend a low threshold for molecular testing in individuals with coronal craniosynostosis alone. One child with plagiocephaly was identified as having the previously described FGFR1 c.899T>C Ile300Thr mutation, which was also present in the unaffected father. Although sagittal craniosynostosis was the single most common referral for unisutural non‐syndromic craniosynostosis (92 individuals), only two mutations were identified: one in a family with autosomal dominant sagittal craniosynostosis and mild Crouzonoid features [reported previously, McGillivray et al., 2005] FGFR2, p.Lys526Glu, and an FGFR2 duplication in a proband with sagittal craniosynostosis and one parent with midface hypoplasia. These studies support the prevailing opinion that molecular testing of FGFR1‐3 and TWIST1 is not warranted in sagittal craniosynostosis unless syndromic in nature or with familial recurrence. These studies support the prevailing opinion that molecular testing of FGFR1‐3 and TWIST1 is not warranted in sagittal craniosynostosis unless syndromic in nature or with familial recurrence. Phenotypes beyond those in the literature were identified in association with a number of specific mutations of importance. These included whole or partial gene deletions of TWIST1 in three patients with Saethre–Chotzen syndrome, only one of whom had intellectual disability. Our recommen- ARTICLE dations for molecular testing have therefore altered to include testing for dosage variance in the TWIST1 gene in patients with a clinical diagnosis of Saethre–Chotzen syndrome who are intellectually normal. The TWIST1 Gly32Ser variant was detected in seven people with phenotypes including unaffected, Saethre–Chotzen and metopic craniosynostosis. This variant has been reported previously as being associated with metopic craniosynostosis and Saethre–Chotzen syndrome [Seto et al., 2007; Foo et al., 2009]. In silico analysis, however suggests that this is not a pathogenic mutation (Condel score 0.001). The FGFR3 Ala391Glu mutation was identified in one individual with a diagnosis of Crouzon with acanthosis nigricans but also in another three individuals without skin manifestations (isolated right unicoronal craniosynostosis, bicoronal craniosynostosis and one with a Crouzon‐like phenotype who was over the age of 10 years). Consideration should be given to testing for this mutation in craniosynostosis in the absence of acanthosis nigricans. The FGFR1 Pro252Arg was identified in two individuals (7.9%) with a clinical diagnosis of Pfeiffer syndrome. Overall, 95% of people tested for Apert syndrome had a canonical mutation identified. Within this cohort, two patients with a clinical diagnosis of Apert syndrome had an Alu element insertion in FGFR2 identified rather than a canonical mutation [Bochukova et al., 2009]. A novel FGFR2 mutation, c.820_824delinsTT p.Val274_Glu275delinsLeu manifested with evolving Crouzon dysmorphology with some features of Costello syndrome, including rugose skin on the extremities. It has been suggested in the literature that mutations in FGFR2 exon 10 are more likely to be associated with a more severe phenotype and the results from this study are consistent with this observation. More detailed review of genotype–phenotype correlations and mutation type has shown that splice site mutations in FGFR2 are also associated strongly with Pfeiffer syndrome versus Crouzon syndrome for non‐splice mutation (P‐value <0.0001, ARTICLE Fisher’s exact test) confirming the findings of Teebi et al. [2002]. Extreme phenotypic variability was observed in a number of families, underscoring the importance of segregation studies in apparently unaffected parents or other first degree relatives and genetic counseling. Mutations which were notable for variability included a novel FGFR2 c.1018T>A p.Tyr340Asn, and a synonymous mutation known to affect FGFR2 mRNA splicing (FGFR2 c.1032G>A p.Ala344Ala). There are relatively few statistical analyses of large cohorts that have related surgical procedures in craniosynostosis to mutation type. The timing and type of transcranial surgery is primarily dependent on the time of referral, symptoms, treatment protocol, the surgeons’ experience and syndromic classification. We were interested in examining if there were any statistical associations between surgical practice and syndromic classification as determined by mutation status. We conducted a retrospective analysis of operative timing of 233 procedures, 208 transcranial and 25 non‐transcranial, performed since 2001. There was borderline difference between the timing of transcranial versus non‐transcranial procedures with non‐transcranial surgery being performed more often in early childhood years. However, the numbers of non‐transcranial procedures recorded were too small to permit further comparisons. The data demonstrate statistically highly significant differences in the type and timing of operative procedures determined by syndromic diagnosis. Of the four syndromic classifications, Saethre–Chotzen patients typically had a transcranial procedure in the first year of life. Transcranial procedures in patients with Crouzon syndrome were performed less frequently in the first year of life, and more frequently in young adulthood. The reverse pattern was seen for Pfeiffer syndrome, with a greater number of surgical procedures performed under 1 year of age. For Apert syndrome, there were fewer procedures performed in young adulthood (SR ¼ 2.72) with the peak period for transcranial operations being between ages 6 and 10 years. AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) SUMMARY AND CONCLUSIONS In summary, we report on the experience of a large referral center involved in clinical assessment, surgical management and mutation testing in craniosynostosis. We have confirmed the clinical utility of a targeted mutation approach to provide cost‐effective services to patients with craniosynostosis. A number of syndromic classifiers are also confirmed to affect surgical management and timing. Although there is a high mutation detection rate in the syndromic craniosynostoses, the etiology of non‐ syndromic craniosynostosis requires additional exploration. Recent studies, which have identified the importance of ERF and TCF12 genes in unisutural and syndromic craniosynostosis, have shown that further increases in mutation detection are possible [Sharma et al., 2013; Twigg et al., 2013]. Next Generation DNA sequencing will enable sensitive and cost‐effective diagnosis for multiple genes in parallel in craniosynostosis patients. As this technology becomes cheaper it will replace the targeted screening approach completely. Given the large number of mutations identified by these technologies, there remains a requirement to describe the clinical phenotype and genotype of highly penetrant mutations in disease genes. The 17 novel mutations identified in this paper will contribute further to the knowledge base of high penetrance mutations in craniosynostosis and so facilitate the recognition and correct classification of mutations in other centers. ACKNOWLEDGMENTS We acknowledge The Australian Cranio‐Maxillo Facial Foundation, Adelaide, for the provision of funds to undertake this research and the cooperation of the families involved. REFERENCES Beare JM, Dodge JA, Nevin NC. 1969. Cutis gyratum, acanthosis nigricans and other congenital anomalies; a new syndrome. Br J Dermatol 81:241–247. 269 Bellus GA, Gaudenz K, Zackai EH, Clarke LA, Szabo J, Francomano CA, Muenke M. 1996. Identical mutations in three different fibroblast growth factor receptor genes in autosomal dominant craniosynostosis syndromes. Nat Genet 14:174–176. Bochukova EG, Roscioli T, Hedges DJ, Taylor IB, Johnson D, David DJ, Deininger PL, Wilkie AO. 2009. Rare mutations of FGFR2 causing Apert syndrome: Identification of the first partial gene deletion, and an Alu element insertion from a new subfamily. Hum Mutat 30:204–211. Foo R, Guo Y, McDonald‐McGinn DM, Zackai EH, Whitaker LA, Bartlett SP. 2009. The natural history of patients treated for TWIST1‐confirmed Saethre–Chotzen syndrome. Plast Reconstr Surg 124:2085–2095. Gilissen C, Arts HH, Hoischen A, Spruijt L, Mans DA, Arts P, van Lier B, Steehouwer M, van Reeuwijk J, Kant SG, Roepman R, Knoers NV, Veltman JA, Brunner HG. 2010. Exome sequencing identifies WDR35 variants involved in Sensenbrenner syndrome. Am J Hum Genet 87:418–423. Glass IA, Chapman S, Hockley A. 1994. A distinct autosomal dominant craniosynostosis brachydactyly syndrome. Clin Dysmorphol 3:215–223. González‐Pérez A, López‐Bigas N. 2011. Improving the assessment of the outcome of nonsynonymous SNVs with a consensus deleteriousness score, Condel. Am J Hum Genet 88:440–449. Gorlin RJ, Cohen MM, Levin LS, editors. 1990. Syndromes of the head and neck. 3rd edition. UK: Oxford University Press. pp 519–539. Graham JM Jr, Braddock SR, Mortier GR, Lachman R, Van Dop C, Jabs EW. 1998. Syndrome of Coronal Craniosynostosis with Brachydactyly and Carpal/Tarsal Coalition due to Pro250Arg Mutation in FGFR3 Gene. Am J Med Genet 77:322–329. Ingersoll RG, Paznekas WA, Tran AK, Scott AF, Jiang G, Jabs EW. 2001. Fibroblast growth factor receptor 2 (FGFR2): Genomic sequence and variations. Cytogenet Cell Genet 94:121–126. Jabs EW, Muller U, Li X, Ma L, Luo W, Haworth IS, Klisak I, Sparkes R, Warman ML, Mulliken JB, Snead ML, Maxson R. 1993. A mutation in the homeodomain of the human MSX2 gene in a family affected with autosomal dominant craniosynostosis. Cell 75:443–450. Jabs EW, Li X, Scott AF, Meyers G, Chen W, Ecles M, Mao J, Charnas LR, Jackson CE, Jaye M. 1994. Jackson‐Weiss and Crouzon syndromes are allelic with mutations in fibroblast growth factor receptor 2. Nat Genet 8: 275–279. Justice CM, Yagnik G, Kim Y, Peter I, Jabs EW, Erazo M, Ye X, Ainehsazan E, Shi L, Cunningham ML, Kimonis V, Roscioli T, Wall SA, Wilkie AO, Stoler J, Richtsmeier JT, Heuzé Y, Sanchez‐Lara PA, Buckley MF, Druschel CM, Mills JL, Caggana M, Romitti PA, Kay DM, Senders C, Taub PJ, Klein OD, Boggan J, Zwienenberg‐Lee M, Naydenov C, Kim J, Wilson AF, Boyadjiev SA. 2012. A genome‐wide association study identifies susceptibility loci for nonsyndromic sagittal craniosynostosis near BMP2 and within BBS9. Nat Genet 44:1360–1364. 270 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS) Kan SH, Elanko N, Johnson D, Cornejo‐Roldan L, Cook J, Reich EW, Tomkins S, Verloes A, Twigg SR, Rannan‐Eliya S, McDonald‐ McGinn DM, Zackai EH, Wall SA, Muenke M, Wilkie AO. 2001. Genomic screening of fibroblast growth‐factor receptor 2 reveals a wide spectrum of mutations in patients with syndromic craniosynostosis. Am J Hum Genet 70:472–486. Krieger E, Koraimann G, Vriend G. 2002. Increasing the precision of comparative models with YASARA NOVA—A self‐ parameterizing force field. Proteins 47: 393–402. McGillivray G, Savarirayan R, Cox TC, Stojkoski C, McNeil R, Bankier A, Bateman JF, Roscioli T, Gardner RJ, Lamandé SR. 2005. Familial scaphocephaly syndrome caused by a novel mutation in the FGFR2 tyrosine kinase domain. J Med Genet 42: 656–662. Meyers GA, Orlow SJ, Munro IR, Przylepa KA, Jabs EW. 1995. Fibroblast growth factor receptor 3 (FGFR3) transmembrane mutation in Crouzon syndrome with acanthosis nigricans. Nat Genet 11:462–464. Moloney DM, Wall SA, Ashworth GJ, Oldridge M, Glass IA, Francomano CA, Muenke M, Wilkie AOM. 1997. Prevalence of Pro250Arg mutation of fibroblast growth factor receptor 3 in coronal craniosynostosis. Lancet 349:1059–1062. Muenke M, Schell U, Hehr A, Robin NH, Losken HW, Schinzel A, Pulleyn LJ, Rutland P, Reardon W, Malcolm S, Winter RM. 1994. A common mutation in the fibroblast growth factor receptor 1 gene in Pfeiffer syndrome. Nat Genet 8:269–274. Muenke M, Gripp KW, McDonald‐McGinn DM, Gaudenz K, Whitaker LA, Bartlett SP, Markowitz RI, Robin NH, Nwokoro N, Mulvihill JJ, Losken HW, Mulliken JB, Guttmacher AE, Wilroy RS, Clarke LA, Hollway G, Ades LC, Haan EA, Mulley JC, Cohen MM Jr, Bellus GA, Francomano CA, Moloney DM, Wall SA, Wilkie AOM, Zackai EH. 1997. A unique point mutation in the fibroblast growth factor receptor 3 gene (FGFR3) defines a new craniosynostosis syndrome. Am J Hum Genet 60:555–564. Passos‐Bueno MR, Serti Eacute AE, Jehee FS, Fanganiello R, Yeh E. 2008. Genetics of craniosynostosis: Genes, syndromes, mutations and genotype–phenotype correlations. Front Oral Biol 12:107–143. Reardon W, Winter RM, Rutland P, Pulleyn LJ, Jones BM, Malcolm S. 1994. Mutations in the fibroblast growth factor receptor 2 gene cause Crouzon syndrome. Nat Genet 8:98– 103. Roscioli T, Flanagan S, Kumar P, Masel J, Gattas M, Hyland VJ, Glass IA. 2000. Clinical findings in a patient with FGFR1 P252R mutation and comparison with the literature. Am J Med Genet 93:22–28. Rutland P, Pulleyn LJ, Reardon W, Baraitser M, Hayward R, Jones B, Malcolm S, Winter RM, Oldridge M, Slaney SF, Poole MD, Wilkie AOM. 1995. Identical mutations in the FGFR2 gene cause both Pfeiffer and Crouzon syndrome phenotypes. Nat Genet 9:173–176. Seto ML, Hing AV, Chang J, Hu M, Kapp‐Simon KA, Patel PK, Burton BK, Kane AA, Smyth MD, Hopper R, Ellenbogen RG, Stevenson K, Speltz ML, Cunningham ML. 2007. Isolated sagittal and coronal craniosynostosis associated with TWIST box mutations. Am J Med Genet Part A 143:678–686. Sharma VP, Fenwick AL, Brockop MS, McGowan SJ, Goos JA, Hoogeboom AJ, Brady AF, Jeelani NO, Lynch SA, Mulliken JB, Murray DJ, Phipps JM, Sweeney E, Tomkins SE, Wilson LC, Bennett S, Cornall RJ, Broxholme J, Kanapin A, 500 Whole‐Genome Sequences (WGS500) Consortium, Johnson D, Wall SA, van der Spek PJ, Mathijssen IM, Maxson RE, Twigg SR, Wilkie AO, 2013. Mutations in TCF12, encoding a basic helix‐ loop‐helix partner of TWIST1, are a frequent cause of coronal craniosynostosis. Nat Genet 45:304–347. Stevenson RE, Ferlauto JG, Taylor HA. 1978. Cutis gyratum and acanthosis nigricans associated with other anomalies: a distinctive syndrome. J Pediatr 92:950–952. Tavtigian SV, Byrnes GB, Goldgar DE, Thomas A. 2008. Classification of rare missense substitutions, using risk surfaces, with genetic‐ and molecular‐epidemiology applications. Hum Mutat 29:1342–1354. Teebi AS, Kennedy S, Chun K, Ray PN. 2002. Severe and mild phenotypes in Pfeiffer syndrome with splice acceptor mutations in exon IIIc of FGFR2. Am J Med Genet 107:43–47. Twigg SRF, Kan R, Babbs C, Bochukova EG, Robertson SP, Wall SA, Morriss‐Kay GM, Wilkie AOM. 2004. Mutations of ephrin‐B1 (EFNB1), a marker of tissue boundary ARTICLE formation, cause craniofrontonasal syndrome. Proc Nat Acad Sci 101:8652– 8657. Twigg SRF, Versnel SL, Nurnberg G, Lees MM, Bhat M, Hammond P, Hennekam RCM, Hoogeboom AJM, Hurst JA, Johnson D, Robinson AA, Scambler PJ, Gerrelli D, Nurnberg P, Mathijssen IMJ, Wilkie AOM. 2009. Frontorhiny, a distinctive presentation of frontonasal dysplasia caused by recessive mutations in the ALX3 homeobox gene. Am J Hum Genet 84: 698–705. Twigg SR, Vorgia E, McGowan SJ, Peraki I, Fenwick AL, Sharma VP, Allegra M, Zaragkoulias A, Sadighi Akha E, Knight SJ, Lord H, Lester T, Izatt L, Lampe AK, Mohammed SN, Stewart FJ, Verloes A, Wilson LC, Healy C, Sharpe PT, Hammond P, Hughes J, Taylor S, Johnson D, Wall SA, Mavrothalassitis G, Wilkie AO. 2013. Reduced dosage of ERF causes complex craniosynostosis in humans and mice and links ERK1/2 signaling to regulation of osteogenesis. Nat Genet 45:308–313. Vissers LE, Cox TC, Maga AM, Short KM, Wiradjaja F, Janssen IM, Jehee F, Bertola D, Liu J, Yagnik G, Sekiguchi K, Kiyozumi D, van Bokhoven H, Marcelis C, Cunningham ML, Anderson PJ, Boyadjiev SA, Passos‐ Bueno MR, Veltman JA, Smyth I, Buckley MF, Roscioli T. 2011. Heterozygous mutations of FREM1 are associated with an increased risk of isolated metopic craniosynostosis in humans and mice. PLoS Genet 7: e1002278. Vriend G. 1990. WHAT IF: A molecular modeling and drug design program. J Mol Graph 8:52–56,29. Wilkie AOM. 1997. Craniosynostosis: Genes and mechanisms. Hum Mol Genet 6:1647– 1656. Wilkie AOM, Slaney SF, Oldridge M, Poole MD, Ashworth GJ, Hockley AD, Hayward RD, David DJ, Pulleyn LJ, Rutland P, Malcolm S, Winter RM, Reardon W. 1995. Apert syndrome results from localized mutations of FGFR2 and is allelic with Crouzon syndrome. Nat Genet 9:165–172. Wilkie AO, Byren JC, Hurst JA, Jayamohan J, Johnson D, Knight SJ, Lester T, Richards PG, Twigg SR, Wall SA. 2010. Prevalence and complications of single‐gene and chromosomal disorders in craniosynostosis. Pediatrics 126:e391–e400.