Microdeletion 9q22.3 syndrome includes metopic craniosynostosis, hydrocephalus, macrosomia, and developmental delay RESEARCH ARTICLE Microdeletion 9q22.3 Syndrome Includes Metopic Craniosynostosis, Hydrocephalus, Macrosomia, and Developmental Delay Eric A. Muller, 1 Swaroop Aradhya, 2 Joan F. Atkin, 3 Erin P. Carmany, 4 Alison M. Elliott, 5 Albert E. Chudley, 5 Robin D. Clark, 6 David B. Everman, 7 Shannon Garner, 3 Bryan D. Hall, 8 Gail E. Herman, 3 Emma Kivuva, 9 Subhadra Ramanathan, 6 David A. Stevenson, 10 David W. Stockton, 4 and Louanne Hudgins 1 * 1 Stanford University, Stanford, California 2 GeneDx, Gaithersburg, Maryland 3 Nationwide Children’s Hospital, Columbus, Ohio 4 Wayne State University and Children’s Hospital of Michigan, Detroit, Missouri 5 Children’s Hospital, Health Sciences Centre, University of Manitoba, Winnipeg, Manitoba, Canada 6 Loma Linda University Health Center, Loma Linda, California 7 Greenwood Genetic Center, Greenwood, South Carolina 8 University of Kentucky, Lexington, Kentucky 9 Royal Devon and Exeter Hospital, Exeter, UK 10 University of Utah School of Medicine, Salt Lake City, Utah Received 13 January 2011; Accepted 27 June 2011 Basal cell nevus syndrome (BCNS), also known as Gorlin syn- drome (OMIM #109400) is a well-described rare autosomal dominant condition due to haploinsufficiency of PTCH1. With the availability of comparative genomic hybridization arrays, increasing numbers of individuals with microdeletions involving this locus are being identified. We present 10 previ- ously unreported individuals with 9q22.3 deletions that include PTCH1. While 7 of the 10 patients (7 females, 3 males) did not meet strict clinical criteria for BCNS at the time of molecular diagnosis, almost all of the patients were too young to exhibit many of the diagnostic features. A number of the patients exhibited metopic craniosynostosis, severe obstructive hydro- cephalus, and macrosomia, which are not typically observed in BCNS. All individuals older than a few months of age also had developmental delays and/or intellectual disability. Only facial features typical of BCNS, except in those with prominent mid- foreheads secondary to metopic craniosynostosis, were shared among the 10 patients. The deletions in these individuals ranged from 352 kb to 20.5 Mb in size, the largest spanning 9q21.33 through 9q31.2. There was significant overlap of the deleted segments among most of the patients. The smallest common regions shared among the deletions were identified in order to localize putative candidate genes that are potentially responsible foreachofthenon-BCNSfeatures.Thesewerea929 kbregionfor metopic craniosynostosis, a 1.08 Mb region for obstructive hydrocephalus, and a 1.84 Mb region for macrosomia. Addi- tional studies are needed to further characterize the candidate genes within these regions. � 2011 Wiley Periodicals, Inc. Key words: PTCH1; 9q22; Gorlin syndrome; basal cell nevus syndrome; basal cell carcinoma syndrome; metopic craniosynos- tosis; chromosomal deletion *Correspondence to: Louanne Hudgins, MD, Stanford University Division of Medical Genetics, 300 Pasteur Drive, H315, Stanford, CA 94305-5208. E-mail: lhudgins@stanford.edu Published online 21 December 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ajmg.a.34216 How to Cite this Article: Muller EA, Aradhya S, Atkin JF, Carmany EP, Elliott AM, Chudley AE, Clark RD, Everman DB,GarnerS,HallBD, Herman GE,Kivuva E, Ramanathan S, Stevenson DA, Stockton DW, Hudgins L. 2012. Microdeletion 9q22.3 syndrome includes metopic craniosynostosis, hydrocephalus, macrosomia, and developmental delay. Am J Med Genet Part A 158A:391–399. � 2011 Wiley Periodicals, Inc. 391 INTRODUCTION Basal cell nevus syndrome (BCNS) is a well described cancer predisposition and multiple malformation syndrome, manifested by a constellation of findings that include lamellar calcification of the falx cerebri at less than 20 years of age, palmar and/or plantar pits, jawkeratocysts,basalcellcarcinomasatlessthan30yearsofage (or more than four in a lifetime), cleft lip and/or cleft palate, vertebral and rib anomalies, pre- or postaxial polydactyly, ocular anomalies, occipitofrontal circumference of greater than 97th centile, and an increased risk of childhood medulloblastomas (primitive neuroectoderm tumor, PNET), lymphomesenteric or pleuralcysts,andovarianorcardiacfibromas[Kimonisetal.,2004]. Variousdysmorphicfacialfeatureshavebeendescribed,including a broad forehead with bossing, hypertelorism, epicanthal folds, midfacial hypoplasia, short, high-set upturned nose, a long and tented philtrum, and coarsened facies. This condition is known to result from autosomal dominant loss of function mutations in the PTCH1geneandfromhaploinsufficiencyofthehumanhomologto the Drosophila patched protein [Hahn et al., 1996; Wicking et al., 1997].WhiletheaverageageofclinicaldiagnosisofBCNSis25years of age, with the increasing use of array-based comparative genomic hybridizationtechnology(CGH),individualswithcontiguousgene deletions that include PTCH1 are being identified at earlier ages, often before manifestation of the characteristic age-dependent BCNS features [Slater et al., 2000]. To date, 27 patients have been reported with 9q22 deletions that either met clinical criteria for diagnosis of BCNS, or include PTCH1 within the deleted interval, excluding intragenic deletions [Sekhon et al., 1982; Ying et al., 1982; Farrell et al., 1991; Robb et al., 1991; Evans et al., 1993; Kroes et al., 1994; Shimkets et al., 1996; Paoloni-Giacobino et al., 2000;L’Hermineetal.,2002;Olivierietal.,2003;Haniffaetal.,2004; Midro et al., 2004; Boonen et al., 2005; Cajaiba et al., 2006; Chen et al., 2006; Redon et al., 2006; Fujii et al., 2007; Nowakowska et al., 2007; de Ravel et al., 2009; Musani et al., 2009; Shimojima et al., 2009; Yamamoto et al., 2009]. Recently, several authors have reported an emerging phenotype separatefromthatofBCNSresultingfrommicrodeletionsof9q22.3 that include PTCH1 [Redon et al., 2006; Shimojima et al., 2009; Yamamoto et al., 2009]. Macrosomia in these individuals has been hypothesized to result from de novo deletions of the paternal allele specifically, possibly due to the loss of one or more as-of-yet unidentified imprinted genes, or possibly PTCH1 itself [Redon et al., 2006; Yamamoto et al., 2009]. Here we report on 10 new patients with 9q22.3 deletions that include PTCH1, of varying ethnic backgrounds and age at diag- nosis, whose deletions were analyzed by array CGH after referral for evaluation by a medical geneticist. We characterize new phenotypic featuresnotconsistentwithBCNS(metopic craniosyntosis, obstruc- tive hydrocephalus, macrosomia, and developmental delay), and map the common areas of overlap among these deletions. MATERIALS AND METHODS Study Design Patient 7 was ascertained by the lead and corresponding authors at an outpatient referral for consultation by a medical geneticist. The other authorswere thencontacted directly for solicitationof similar patients for inclusion in a retrospective case review series study, after approval of the study and consent forms by the Stanford Institutional Review Board (IRB). The other nine affected individ- uals had been previously evaluated by one or more of the other authors after referral from a medical geneticist evaluation. Clinical and molecular diagnosis of BCNS was confirmed by detection of an interstitial 9q22.3 deletion that included PTCH1 by array-based CGH, including BAC array, 44K oligonucleotide array, 105K oligonucleotide array, and/or SNP array; some individuals had more than one of these analyses. The only study inclusion criterion was a confirmed molecular diagnosis by one of the described methods in a previously unreported individual. Medical records regarding the affected individual’s phenotype, including clinic notes by the authors, cytogenetic study reports, and facial photographs if explicitly authorized, were sent by each physician for compilation and review by the lead author. In compiling the phenotypic features for each patient, macrosomia was defined as both height or length and weight of greater than 95th centile. Analysis Each individual’s reported deletion breakpoints were put into the UCSC Genome Browser (http://genome.ucsc.edu/) NCBI36/hg18 March 2006 build for analysis, in accordance to the human genome assembly used in the cytogenetic report for each patient [Kent et al., 2002]. The isolated regions were then converted to the current GRCh37/hg19 February 2009 human genome assembly to obtain the revised breakpoints, using tools available at the Genome Browser site. The revised, current build breakpoints were then input into the NCBI MapViewer website for gene annotation using human genome build 37.2 (http://www.ncbi.nlm.nih.gov/ mapview/). Basepairs described in this report are for the NCBI36/ hg18 March 2006 build. RESULTS The 10 affected individuals included 7 females and 3 males, who ranged in age at time of initial geneticist evaluation from newborn to 7 years. Their ages at the time of confirmed array-based molec- ular diagnosis spanned from less than 1 month to 13 years. Gestational ages at birth ranged from 37 to 41 weeks. Each of these individuals was the only affected member in his or her family, and the deletions were confirmed as de novo in the nine for whom parental samples were obtained. The ethnic background was Euro- peanCaucasian(8),oneofwhomwasMennonite,Mexican(1),and Chinese (1). These findings are summarized in Table I. Using the previously proposed diagnostic criteria of two major features, or one major plus two minor features, only three indi- viduals (Patients 4, 5, and 6) met a clinical diagnosis of BCNS [Evansetal.,1993;Slateretal.,2000].Eachpersonexhibitedsomeof the characteristic features of BCNS, which included absolute mac- rocephaly (6/10), rib and/or vertebral anomalies (5/8), ocular malformations (3/10), numerous basal cell nevi (3/10), mandibular odontogenic keratocysts (3/10), Sprengel deformity (3/10), cardiac fibromas (2/8), polydactyly (2/10), palmoplantar pits (2/10), and 392 AMERICAN JOURNAL OF MEDICAL GENETICS PART A palatal clefting (2/10), among other features. Features not charac- teristic of BCNS were metopic craniosynostosis, hydrocephalus, macrosomia, and developmental delay with or without intellectual disability. Trigonocephaly, resulting from metopic craniosynosto- sis, was present in four individuals (Patients 1, 5, 7, and 8), each of whom underwent corrective surgery. Obstructive hydrocephalus was apparent in five individuals (Patients 2–5, 9), each of whom required ventriculo-peritoneal shunting. One child, Patient 6, had TABLE I. Summary of Patient Data Features Patient 1 2 3 4 5 6 7 8 9 10 Gender M F F M F F M F F F Presenting age Birth 10 Days 2 Weeks 9 y Birth 9 mo 7 mo Birth 5 y 7 y Race/Ethnicity Caucasian Caucasian Caucasian Mennonite Caucasian Chinese Mexican Caucasian Caucasian Caucasian Diagnosis age 2 weeks 1 y 1 mo 9 y 13 y 2 y 2 y 2 mo 11 y 7 y Last exam age 4 mo 5 y 9.5 mo 22 y 16 y 3 y 3 y 7 mo 13 mo N/A N/A Deletion (Mb) 20.5 10.85 9.85 8.28 8.07 4.5 2.03 1.84 1.08 0.352 Metopic fusion þ � � � þ � þ þ � � Hydrocephalus � þ þ þ þ � � � þ � Macrosomia � � � � � þ þ/� þ � � Delay ? Global Motor a Global Global Global Global Motor Motor Motor Intellect impaired ? þ ? þ þ ? ? ? � � Molecular array BAC BAC, 105K 105K SNP 105K BAC, 105K BAC, 105K 105K 105K, MLPA 44K BCNS FEATURES Craniofacial Bossing � þ þ þ � þ þ � � þ Coarse face þ þ þ � þ � � � � � Synophrys þ � � � � � � � � � Epicanthal folds þ þ � � þ þb þ þ � � Palpebral slanting � Down � � Up Down � � Down Up Cleft Palate � � Uvula � � � � � � Short nose þ þ � þ þ þ � þ � � Philtrum: Long þ þ � þ þ � � þ � � Tented þ þ � � þ � � þ � � Ears: low, angled þ � þ þ þ � � þ þ þ Abnormal dentition ? � � � � � � þ þ � Jaw keratocysts � � � þ þ � � � þ � Macrocephaly � þ � þ þ þ þ þ � þ Ocular anomalies � � Nerve � � � � Micro Retinal � Skeletal Rib anomaly þ ? � þ � þ þ ? þ � Vertebral anomaly þ ? � � � þ � ? þ � Sprengel deformity � � � þ � þ � � þ � Pectus deformity � � þ þ � � � � � � (Kypho)scoliosis þ þ � � � þ � � þ � Polydactyly � � � � � þ � þ � � Syndactyly � � � � � � � þ � � Oligodactyly � � � þ þc � � � � � Cardiac tumor RV � � � RA � � � ? ? Dermatological Basal cell Nevi � � � þ � þ þ � � � Palmoplantar Pits � � � � � þ � � � þ Neurological Hypotonia þ þ þ � þ þ � � � � þ, present; �, absent; ?, unknown or not assessed; 44K, 44K-oligonucleotide array; 105K, 105K-oligonucleotide array; BAC, bacterial artificial chromosome array; micro, microphthalmia and optic nerve hypoplasia; MLPA, multiplex ligation-dependent probe amplification; N/A, not applicable; nerve, optic nerve anomaly; retinal, abnormal retinal myelination; RA, right atrial; RV, right ventricular; SNP, single nucleotide polymorphism array. a Not assessed for speech or other delays. b Epicanthal folds were present, but were also consistent with patient’s ethnicity. c Bilateral thumb hypoplasia was present. MULLER ET AL. 393 mildventricularenlargement thatdidnotrequire shunting.Macro- somia was present in two individuals (6, 8), each of whom also had macrocephaly. One child (Patient 7) had a height between the 90th and 95th centiles, and a weight at the 98th centile. As this did not meet our strict definition of macrosomia, his deletion was excluded from the overlap region analysis for macrosomia. In all eight individuals who were older than a few months of age and who could be fully assessed, developmental delays were apparent. Only motor delays were seen in the children with the smallest deletions, which in some cases resolved over time. Global motor, speech/ language, and social/behavioral impairments were observed in those with the larger deletions. Intellectual disability was apparent in the three globally delayed patients who were followed to school age and older. Hypotonia was present in half of the individuals (Table I). Various dysmorphic facial features were present in a number of patients, and included a broad forehead with bossing, vertical forehead creases, epicanthal folds, angulated palpebral fissures, micrognathia, and low-set and posteriorly rotated ears. Noted among many of the patients were a short nose with the appearance of being ‘‘high set,’’ and a long and tented philtrum (Table I; Fig. 1). The facial features also coarsened over time in two individuals (Patients 2, 8; Fig. 2). Features present in some members of our population were synophrys, hypertelorism, partial corpus callosum dysgenesis, Chiari I malformation, seizures, joint hypermobility, camptodac- tyly, ataxia, short stature, broad/webbed neck, and inguinal hernias (data not shown). One child (6) had unilateral renal cysts, mild ipsilateral ureterectasis that resolved spontaneously, and contrala- teral mild renal pelviectasis. None of the other nine individuals had any documented renal anomalies. All of our reported individuals except Patient 9 had G-banded chromosomal analysis. The deletions of Patients 1, 3, and 4 were visibleonhigh-resolutionanalysis.Theothershadlower-resolution chromosomal analysis, and their deletions were not visible micro- scopically (data not shown). Overall, the deletion sizes ranged from 352 kb, containing only PTCH1 and part of FANCC, to 20.5 Mb, spanning 273 genes, as illustrated (Fig. 3). Individual 3 also had a second 44 Kb deletion located at 9q21.3, which contained no genes (data not shown). The smallest region of deletion overlap for the five affected individuals with metopic craniosynostosis was 929 kb, between bases 96,771,450 and 97,700,488, and which contained FIG. 1. Five individuals with interstitial 9q22.3 deletions. Please note the short nose and long philtrum that is best observed in patients 1, 3, and 8, but isalsopresentinpatient4,andthelowsetandposteriorlyrotatedearsinallofthepatients.Tentingofthephiltrumisalsopresentinpatients1and8. 394 AMERICAN JOURNAL OF MEDICAL GENETICS PART A FIG. 2. Age progression in individuals with interstitial 9q22.3 deletions. A. Patient 2 at age 6 months and 19 months (top row, left-to-right), and at 34 months, 4 years, and 5 years (bottom row, left-to-right). Age progression in individuals with interstitial 9q22.3 deletions. B. Patient 5 at age 5 months (top and middle rows), 11 years (bottom left), and 16 years (bottom right). Again note the short nose and long, tented philtrum in both patients. FIG. 3. Schematicrepresentationofthedeletionsinaffectedindividuals.Thedeletionsforeachaffectedindividual, listedontheleft,arerepresentedto scale by horizontal bars of various colors, in comparison to a chromosome 9q ideogram along the top. Bars composed of two colors indicate the presence of both features. The relative position and size of PTCH1 is represented by a vertical bar. Listed deletion breakpoints are according to the NCBI 36/hg18 build. MULLER ET AL. 395 15annotatedgenesother thanPTCH1.Therewasacommonareaof deletionoverlapof1.08 Mbamongpatientswithobstructivehydro- cephalus that contained 17 genes other than PTCH1. This region localized to bases 96,848,332 through 97,930,074. The smallest region of deletion overlap for the two patients with macrosomia was 1.8 Mb, located between bases 96,771,450 and 98,619,830. This region contained 30 genes other than PTCH1. DISCUSSION We have expanded the phenotype of individuals with interstitial 9q22 deletions that contain PTCH1 to include metopic craniosy- nostosis, obstructive hydrocephalus, and macrosomia, which are separate from the typical features of BCNS. This would suggest that there are genes other than PTCH1 in the deleted region that when deleted contribute to these conditions. The common area of overlap among the patients with metopic craniosynostosis, resulting in trigonocephaly, was between base- pairs 96,771,450 and 97,700,488. There are 15 genes other than PTCH1 within this region. These are FANCC, encoding the Fanconi anemia complementation group C protein, open reading frames that encode a zinc aminopeptidase (c9orf3), a putative repair/recombination helicase (c9orf102), and a nasopharyngeal carcinoma-associated protein (c9orf130), four pseudogenes, two hypothetical RNA-encoding genes (LOC100507319, LOC100507346), one gene that is expressed in human heart and brain and encodes a hypothetical protein of unknown function (LOC100506667), and four microRNA genes. While there are no obvious candidate genes among those listed, it is possible that the deleted region may include one gene whose haploinsufficiency is but one of many factors that influence the development of metopic craniosynostosis in affected individuals, perhaps explaining why every individual who shared our mapped minimal region for metopic craniosynostosis did not demonstrate that finding. The presence of metopic craniosynos- tosis in a substantial percentage of all reported individuals with 9q22.3 microdeletion is interesting. We do suggest that 9q22.3 microdeletion be included in the differential diagnosis for individ- uals with metopic craniosynostosis in the absence of multiply affected sutures, and as such, CGH studies may be indicated. In comparison, the commonly deleted region among individuals with obstructive hydrocephalus was located between 96,848,332 and 97,930,074. Within this region were the same three open reading frames, four microRNA genes, FANCC, hypothetical pro- tein gene, and one of the two hypothetical non-protein encoding RNA genes which were also present in the deletions of individuals with metopic craniosynostosis. Unique to this region was another pseudogene, a non-protein coding RNA gene, and an additional hypothetical RNA-encoding gene. Again, there are no obvious candidate genes among them. We note that asymmetric ventricles, dilated ventricles, and dilation due to cerebral atrophy have been reported in the literature on patients with the clinical diagnosis of BCNS [Chen et al., 2006]. We cannot determine whether all of these individuals have inter- stitial deletions that contain PTCH1, as the diagnosis of BCNS was purely clinical before the genetic etiology was identified [Hahn et al., 1996]. Similarly, some of the 27 previously described indi- viduals with known interstitial 9q22 deletions that contain PTCH1 by mapping or by BCNS diagnosis have had ventricular dilation of varying degrees [Robb et al., 1991; Kroes et al., 1994; Midro et al., 2004; Boonen et al., 2005; Cajaiba et al., 2006; Chen et al., 2006; Redon et al., 2006; Fujii et al., 2007], or like many of our newly reported patients, hydrocephalus requiring ventriculo-peritoneal shunting [Farrell et al., 1991; Shimkets et al., 1996]. Without the higher resolution mapping available with current technology, we cannot refine the critical regions for those patients beyond those large regions achieved by linkage analysis or BAC array-based methods utilized in those studies. We do expect the critical region to overlap with our proposed minimal intervals, however, due to the large size of the reported deletions in the older case reports. In two of the most recently described patients, mild dilation of one or more ventricles was present, and linked to the presence of a space- occupying lesion [Shimojima et al., 2009; Yamamoto et al., 2009]. We do not consider this to be obstructive hydrocephalus of a nature similar to that in our described patients who required shunting, though one of our patients (6) did have mild enlargement of the ventricle and cerebral sulci (Table II). Overall, a higher proportion of our newly reported individuals had more severe forms of ventricular dilation requiring intervention than those reported previously, whereas the earlier reports described a higher frequency of less severe forms. This may be the result of the more severely affected individuals being brought to medical attention at younger ages in response to rapidly increasing head size, and/or seizures, prompting additional evaluations for other abnormal phenotypic features, and thus allowing diagnosis of 9q22.3 microdeletion earlier than would otherwise be expected. Alternatively, milder forms may only be ascertained incidentally as part of a brain imaging evaluation for developmental delay or as part of the neuroimaging for BCNS, may be generally underreported, or may be missed altogether because such imaging is sometimes difficult in developmentally delayed or impaired children who are behaviorally noncompliant during the study. We suggest that brain imaging be performed as part of the initial evaluation for 9q22.3microdeletion, andsurveillancemonitoring ofhead circum- ference be performed at all clinical visits (Table III). We described two individuals with macrosomia, as pre-defined by height or length and weight each at the 95th centile or greater. The deletions of these children contained the common region of basepairs 96,771,450 through 98,619,830. In addition to the pre- vious 18 genes (other than PTCH1) described for the regions common among affected individuals with metopic craniosynos- tosis and among those with hydrocephalus, there were 12 genes that were unique to this region. These included three pseudogenes, a hypothetical RNA-encoding gene, and an open reading frame (c9orf21) that encodes a protein with peroxiredoxin-like domains. Also among the genes unique to this common deletion were the genes HSD17B3, which encodes a 17-hydroxy-steroid dehydrogen- ase; SLC35D2, which encodes an endoplasmic reticulum/Golgi apparatus nucleotide sugar transporter; and, HABP4, encoding an intracellular hyaluronan binding protein. Perhaps most inter- estingly, also located within the region are CDC14B, encoding the human homolog of a yeast cell division cycle protein, and ZNF510 and ZNF782, which encode novel zinc finger-containing proteins that have been implicated in association studies with regard to modulation of stature specifically within the Chinese population 396 AMERICAN JOURNAL OF MEDICAL GENETICS PART A [Lei et al., 2009]. Of note, one of our two reported individuals with macrosomia was Chinese. Previous case reports have focused on correlating overgrowth in individuals with 9q22.3 deletions that include PTCH1 with loss of the paternal allele [Redon et al., 2006; Shimojima et al., 2009]. The inference is that loss of one or more imprinted genes within the region, perhaps even PTCH1, is a cause of macrosomia. In our review of the 27 previously reported affected individuals, six had macrosomia [Cajaiba et al., 2006; Chen et al., 2006; Redon et al., 2006; Shimojima et al., 2009; Yamamoto et al., 2009]. Notably, this number differs from those reviewed and reported by Shimojima et al., as we included an individual with hemihyperplasia, and we excludedthethirdpatientofFarrelletal.[Farrelletal.,1991;Cajaiba et al., 2006]. This latter child had a birth length and weight of 90th and 97th centiles, respectively, and her length had decreased to 75th centile by 7 months of age. Of the six individuals with macrosomia, TABLE III. Suggested Evaluations After a Diagnosis of 9q22.3 Microdeletion Clinical concern Evaluation strategy Hydrocephalus, ventricular dilation Brain imaging Complete neurological evaluation Cerebral tumors a Surveillance monitoring of head circumference Macrosomia, hemihyperplasia Surveillance monitoring of growth parameters Metopic craniosynostosis Complete craniofacial assessment Carefully consider cost versus benefit of radiographic imaging Developmental delay Microarray/CGH of high resolution to confirm extent of deletion Formal developmental assessment Intellectual impairment Childhood intervention services referrals Renal anomalies Renal ultrasound Ovarian fibroma a Abdominal/pelvic ultrasound Basal cell carcinoma a Periodic complete dermatological assessment Avoidance of ionizing and UV radiation exposure Cardiac anomalies, fibroma a Echocardiogram Ocular anomalies a Ophthalmological evaluation. Skeletal anomalies a Carefully consider cost versus benefit of radiographic imaging Odontogenic keratocysts a CGH, comparative genomic hybridization. a Indicates a feature of BCNS. TABLE II. Comparison of Patients With Interstitial 9q22.3 Deletions Non-BCNS features Affected individuals This study Previous studies Total cumulative Number Percent (%) Number Percent (%) Number Percent (%) Metopic craniosynostosis 4/10 40 4/27 14.8 8/37 21.6 Ventricular dilation a Hydrocephalus 5/10 50 2/27 7.4 7/37 18.9 Mild dilation 1/10 10 8/25 b 32 9/35 25.7 Macrosomia c 2/10 20 6/27 22 8/37 21.6 Developmental delay Motor only 3/8 d 37.5 7/22 e 31.8 10/30 33.3 Speech only 0/8 0 0/22 0 0/30 0 Global 5/8 62.5 14/23 e 60.8 19/31 61.3 Intellectual disability 3/5 60 14/23 e 60.8 17/28 60.7 a Hydrocephalus refers to severe and obstructive ventricular dilation that required ventriculo-peritoneal shunting, whereas mild dilation did not, and may have been asymmetric, or of only one ventricle. b Excludes one case caused by cerebral atrophy and two cases caused by space-occupying masses. c Defined as both height (length) and weight >95 percentile. d Excludes patients with motor delay, but who were not old enough to assess speech and for other impairments. e Excludes patients without clear description of delays and/or impairment. MULLER ET AL. 397 five had lost the paternal allele, though Yamamoto et al. did not report the parent of origin for their patient’s allele [Farrell et al., 1991; Cajaiba et al., 2006; Chen et al., 2006; Redon et al., 2006; Shimojima et al., 2009; Yamamoto et al., 2009]. Three other individuals withlossofthepaternalallelehadnormalorunreported growth parameters [Shimkets et al., 1996; Olivieri et al., 2003; Fujii et al., 2007]. In contrast, of the four people who were known to have lost the maternal allele, one was reported as having normal growth, one was reported as having small stature, and the other two did not havereportedgrowth descriptions[Yingetal.,1982;Shimketsetal., 1996; Midro et al., 2004; Fujii et al., 2007; Shimojima et al., 2009; Yamamoto et al., 2009]. We did not determine the parental origin for the deleted allele in our reported individuals, as these studies were not originally performed for any of the individuals on whom parental samples were available, and were outside of the scope permitted by our institutional IRB-approved protocol and consent for this retrospective study. In all of the previously reported individuals with macrosomia for whom we could identify the breakpoints from the published report, each shared the smallest commonly deleted region identified in our two children with macrosomia, though again, comparison with the earlier studies before the use of array CGH technology is difficult [Chen et al., 2006; Redon et al., 2006; Shimojima et al., 2009]. Based on all of the reported patients to date, there is no consistent evidence for imprinted genes located around or including PTCH1, as there do not seem to be two distinct (opposite) phenotypes, each segregating with loss of a particular parental allele, especially with respect to growth. While we cannot definitively exclude the possibility of candidate genes within the commonly deleted inter- val, or their downstream effectors, being imprinted at this time, there is no uniparental disomy 9 phenotype described the in the medical literature [Wilkinson et al., 1996; Slater et al., 2000]. We do feel that additional study with regard to parent of origin for the deleted alleles is warranted, and also suggest consideration of SNP or array CGH as part of the evaluation of individuals with macro- somia or generalized overgrowth. There is no evidence at this time to determine whether routine surveillance for abdominal tumors in individuals with 9q22.3 microdeletion and macrosomia or hemi- hyperplasia is warranted. It is expected that individuals with 9q22.3 microdeletion have the same or similar risk for tumor types that are seen with increased frequency in individuals with BCNS, though there is limited information available, given how few affected individuals have been reported to date. In all eight of our affected individuals who were of sufficient age and could be adequately assessed, developmental delays and/or intellectual disability was present. The smallest deletion, containing only PTCH1 and a portion of FANCC, was present in individual 10, who exhibited motor milestone delays. It is likely that her delays are related to PTCH1 haploinsufficiency alone. In our reported indi- viduals with deletions of approximately 2 Mb and larger, global impairment, including impairment of speech and language, and in the older individuals, intellectual disability, were present in addi- tion to motor developmental delay. This phenomenon is not unique to the specific genes within the deleted intervals, as numer- ous contiguous gene deletion syndromes show more severe impair- ment with increasing deletion sizes [Slavotinek, 2008; Musani et al., 2009]. Our most severely affected person overall, with an approx- imate 20.5 Mb deletion, was too young to be fully assessed, but was noted to have hypotonia and failure to thrive, and died in infancy of complicationsofarespiratoryillness.Hadhelived,itislikelythathe would have had significant developmental and intellectual disabil- ities in the future. The majority of our reported individuals were female. While equal numbers of affected males and females are expected, based on the autosomal nature of the chromosome deletion, we cannot explain this finding by any systemic ascertainment bias. Of the 27previouslydescribedpeoplewith9q22deletionsthatcauseBCNS or include PTCH1, there were 14 males, 12 females, and 1 sex unreported, showing no sex-related skewing. Overall, as summarized in Table II, 40% of our population demonstrated metopic craniosynostosis, compared to approxi- mately 15% of the previously reported 27 individuals with inter- stitial 9q22 deletions. In our reported affected individuals, half had obstructive hydrocephalus, compared to approximately 7% of those who have been previously reported. In contrast, almost a third of the previously reported individuals presented with milder forms, whereas only one child (10%)in our study did.Similarto the previous reports, 20% of our population demonstrated macro- somia of greater than 95th centile for height (or length) and for weight. Developmental delay and/or irreversible intellectual impairment was present to a similar degree both in our affected individuals, and in those in the previous case reports, though direct comparison is difficult because of limited data and the widely variable ages of all of the patients. It is unclear as to the mechanism which predisposes the 9q22.3 region to deletion. In silico analysis using the RepMask 3.27 tool availableontheUCSCGenomeBrowser(http://genome.ucsc.edu/) identifiednumerousSINEs,largeLINEs,andLTRsflankingPTCH1 and adjacent genes in the region, which potentially could result in recombinatory deletion and duplication events. Perhaps reflective of a reciprocal recombination event to the 9q22.3 microdeletion, a mother and child with the shared phenotype of microcephaly and mild developmental delay was each reported to have a 360 kb duplication of the 9q22.3 region that included PTCH1 and exon 1 of FANCC [Derwinska et al., 2009]. Insummary,wedescribe10additionalindividualswithdeletions involving 9q22 that include the PTCH1 gene. The findings in these 10 people broaden the phenotype to include metopic craniosynos- tosis, hydrocephalus, macrosomia, and developmental delay with intellectualdisability.Whilemanyoftheaffectedindividualsshared similar dysmorphic features that have been previously described in patients with BCNS, no other unifying characteristic facial phenotype was recognized, except with respect to that resulting from metopic craniosynostosis. These findings have implications for clinical care in individuals with 9q22.3 microdeletion (summarized in Table III), as well as for identifying genes that contribute to the expanded phenotype. Further studies are neces- sarytofurtherrefinethesmallestareasofoverlapinordertoidentify the genes implicated in these findings. REFERENCES Boonen SE, Stahl D, Kreiborg S, Rosenberg T, Kalscheuer V, Larsen LA, Tommerup N, Brondum-Nielsen K, Tumer Z. 2005. Delineation of an 398 AMERICAN JOURNAL OF MEDICAL GENETICS PART A interstitial 9q22 deletion in basal cell nevus syndrome. Am J Med Genet A 132A:324–328. Cajaiba MM, Bale AE, Alvarez-Franco M, McNamara J, Reyes-Mugica M. 2006. Rhabdomyosarcoma, Wilms tumor, and deletion of the patched gene in Gorlin syndrome. Nat Clin Pract Oncol 3:575–580. Chen CP, Lin SP, Wang TH, Chen YJ, Chen M, Wang W. 2006. Perinatal findings and molecular cytogenetic analyses of de novo interstitial deletion of 9q (9q22.3–>q31.3) associated with Gorlin syndrome. Prenat Diagn 26:725–729. de Ravel TJ, Ameye L, Ballon K, Borghgraef M, Vermeesch JR, Devriendt K. 2009. Early detection of chromosome 9q22.32q31.1 microdeletion and the nevoid basal cell carcinoma syndrome. Eur J Med Genet 52:145–147. Derwinska K, Smyk M, Cooper ML, Bader P, Cheung SW, Stankiewicz P. 2009. PTCH1 duplication in a family with microcephaly and mild developmental delay. Eur J Hum Genet 17:267–271. Evans DG, Ladusans EJ, Rimmer S, Burnell LD, Thakker N, Farndon PA. 1993. Complications of the naevoid basal cell carcinoma syndrome: Results of a population based study. J Med Genet 30:460–464. Farrell SA, Siegel-Bartelt J, Teshima I. 1991. Patients with deletions of 9q22q34 do not define a syndrome: Three case reports and a literature review. Clin Genet 40:207–214. Fujii K, Ishikawa S, Uchikawa H, Komura D, Shapero MH, Shen F, Hung J, Arai H, Tanaka Y, Sasaki K, Kohno Y, Yamada M, Jones KW, Aburatani H, Miyashita T. 2007. High-density oligonucleotide array with sub- kilobase resolution reveals breakpoint information of submicroscopic deletions in nevoid basal cell carcinoma syndrome. Hum Genet 122: 459–466. Hahn H, Wicking C, Zaphiropoulous PG, Gailani MR, Shanley S, Chi- dambaram A, Vorechovsky I, Holmberg E, Unden AB, Gillies S, Negus K, Smyth I, Pressman C, Leffell DJ, Gerrard B, Goldstein AM, Dean M, Toftgard R, Chenevix-Trench G, Wainwright B, Bale AE. 1996. Muta- tionsofthehumanhomologofDrosophilapatchedinthenevoidbasalcell carcinoma syndrome. Cell 85:841–851. Haniffa MA, Leech SN, Lynch SA, Simpson NB. 2004. NBCCSsecondary to an interstitial chromosome 9q deletion. Clin Exp Dermatol 29:542–544. Kent WJ, Sugnet CW, Furey TS, Roskin KM, Pringle TH, Zahler AM, Haussler D. 2002. The human genome browser at UCSC. Genome Res 12:996–1006. Kimonis VE, Mehta SG, Digiovanna JJ, Bale SJ, Pastakia B. 2004. Radio- logical features in 82 patients with nevoid basal cell carcinoma (NBCC or Gorlin) syndrome. Genet Med 6:495–502. KroesHY,Tuerlings JH, Hordijk R,Folkers NR, tenKate LP.1994. Another patient with an interstitial deletion of chromosome 9: Case report and a review of six cases with del(9)(q22q32). J Med Genet 31:156–158. LeiSF, YangTL,TanLJ,ChenXD, GuoY,Guo YF,ZhangL,Liu XG,YanH, Pan F, Zhang ZX, Peng YM, Zhou Q, He LN, Zhu XZ, Cheng J, Liu YZ, Papasian CJ, Deng HW. 2009. Genome-wide association scan for stature in Chinese: Evidence for ethnic specific loci. Hum Genet 125:1–9. L’Hermine AC, Aboura A, Simon-Bouy B, Robin F, Audibert F, Strouk N, Capron F, Frydman R, Tachdjian G. 2002. Female pseudohermaphrodi- tism in a fetus with a deletion 9(q22.2q31.1). Prenat Diagn 22:652–655. Midro AT, Panasiuk B, Tumer Z, Stankiewicz P, Silahtaroglu A, Lupski JR, Zemanova Z, Stasiewicz-Jarocka B, Hubert E, Tarasow E, Famulski W, Zadrozna-Tolwinska B, Wasilewska E, Kirchhoff M, Kalscheuer V, Michalova K, Tommerup N. 2004. Interstitial deletion 9q22.32-q33.2 associated withadditional familialtranslocation t(9;17)(q34.11;p11.2)in a patient with Gorlin-Goltz syndrome and features of Nail-Patella syndrome. Am J Med Genet A 124A:179–191. Musani V, Cretnik M, Situm M, Basta-Juzbasic A, Levanat S. 2009. Gorlin syndrome patient with large deletion in 9q22.32-q22.33 detected by quantitative multiplex fluorescent PCR. Dermatology 219:111–118. Nowakowska B, Kutkowska-Kazmierczak A, Stankiewicz P, Bocian E, Obersztyn E, Ou Z, Cheung SW, Cai WW. 2007. A girl with deletion 9q22.1-q22.32 including the PTCH and ROR2 genes identified by genome-wide array-CGH. Am J Med Genet A 143A:1885–1889. Olivieri C, Maraschio P, Caselli D, Martini C, Beluffi G, Maserati E, Danesino C. 2003. Interstitial deletion of chromosome 9, int del(9)- (9q22.31-q31.2), including the genes causing multiple basal cell nevus syndrome and Robinow/brachydactyly 1 syndrome. Eur J Pediatr 162: 100–103. Paoloni-Giacobino A, Floris E, Dahoun SP. 2000. Fetus with a 9q22q34 interstitial deletion and hygroma. Prenat Diagn 20:855–856. Redon R, Baujat G, Sanlaville D, Le Merrer M, Vekemans M, Munnich A, Carter NP, Cormier-Daire V, Colleaux L. 2006. Interstitial 9q22.3 micro- deletion: Clinical and molecular characterisation of a newly recognised overgrowth syndrome. Eur J Hum Genet 14:759–767. Robb LJ, Vekemans M, Richter A, Meagher-Villemure K, Neilsen K, der Kaloustian K. 1991. Interstitial deletion of the long arm of chromsome 9. Am J Hum Genet 49:274. Sekhon G, Herman J, Neu J, Sun J. 1982. Interstitial deletion of the 9q22q32 region causes a distinct syndrome. Unpublished abstract cited in Farrell SA, Siegel-Bartelt J, Teshima I. 1991. Patients with deletions of 9q22q34 do not define a syndrome: Three case reports and a literature review. Clin Genet 40:207–214. Shimkets R, Gailani MR, Siu VM, Yang-Feng T, Pressman CL, Levanat S, Goldstein A, Dean M, Bale AE. 1996. Molecular analysis of chromosome 9q deletions in two Gorlin syndrome patients. Am J Hum Genet 59: 417–422. Shimojima K, Adachi M, Tanaka M, Tanaka Y, Kurosawa K, Yamamoto T. 2009. Clinical features of microdeletion 9q22.3 (pat). Clin Genet 75: 384–393. Slater HR, Ralph A, Daniel A, Worthington S, Roberts C. 2000. A case of maternal uniparental disomy of chromosome 9 diagnosed prenatally and the related problem of residual trisomy. Prenat Diagn 20:930–932. Slavotinek AM. 2008. Novel microdeletion syndromes detected by chro- mosome microarrays. Hum Genet 124:1–17. Wicking C, Shanley S, Smyth I, Gillies S, Negus K, Graham S, Suthers G, Haites N, Edwards M, Wainwright B, Chenevix-Trench G. 1997. Most germ-line mutations in the nevoid basal cell carcinoma syndrome lead to a premature termination of the PATCHED protein, and no genotype–phenotype correlations are evident. Am J Hum Genet 60:21–26. Wilkinson TA,James RS,Crolla JA, Cockwell AE, Campbell PL, Temple IK. 1996. A case of maternal uniparental disomy of chromosome 9 in association with confined placental mosaicism for trisomy 9. Prenat Diagn 16:371–374. Yamamoto K, Yoshihashi H, Furuya N, Adachi M, Ito S, Tanaka Y, Masuno M, Chiyo H, Kurosawa K. 2009. Further delineation of 9q22 deletion syndrome associated with basal cell nevus (Gorlin) syndrome: Report of two cases and review of the literature. Congenit Anom (Kyoto) 49:8–14. Ying KL, Curry CJ, Rajani KB, Kassel SH, Sparkes RS. 1982. De novo interstitial deletion in the long arm of chromosome 9: A new chromo- some syndrome. J Med Genet 19:68–70. MULLER ET AL. 399