Comparison of Clinical-Radiological and MolecularFindings in Hypochondroplasia

Chiara Prinster,1,2* Paola Carrera,3 Maurizia Del Maschio,4 Giovanna Weber,2 Mohamad Maghnie,6Maria Cristina Vigone,1,2 Stefano Mora,1 Giorgio Tonini,7 Franco Rigon,8 Giampiero Beluffi,5Francesca Severi,6 Giuseppe Chiumello,2 and Maurizio Ferrari3

1Laboratory of Pediatric Endocrinology, Scientific Institute H San Raffaele, University of Milan, Milan, Italy2Department of Pediatrics, Scientific Institute H San Raffaele, University of Milan, Milan Italy3Laboratory of Clinical Molecular Biology, Scientific Institute H San Raffaele, University of Milan, Milan, Italy4Department of Radiology, Scientific Institute H San Raffaele, University of Milan, Milan, Italy5Section of Pediatric Radiology, Department of Radiodiagnosis, University of Pavia, IRCCS Policlinico S. Matteo,Pavia, Italy

6Pediatric Clinic, University of Pavia, IRCCS Policlinico S. Matteo, Pavia, Italy7Department of Pediatrics, IRCCS per l’Infanzia Burlo-Garofolo, University of Trieste, Trieste, Italy8Department of Pediatrics, University of Padua, Padua, Italy

Hypochondroplasia is an autosomal domi-nant skeletal dysplasia characterized bydisproportionate short stature. A mutation(N540K) in the fibroblast growth factor re-ceptor 3 (FGFR3) gene was described insome patients with this condition. The aimsof the study were to identify the frequencyof the FGFR3 gene mutation, to define thesalient clinical and radiological abnormali-ties of the affected subjects, and to verifythe contribution of molecular findings tothe clinical and radiological definition ofhypochondroplasia.

Based on the most common radiologicalcriteria, we selected 18 patients with a phe-notype compatible with hypochondroplasia.Height, sitting height, and cranial circum-ference were measured in all patients. Ra-diographs of the lumbar spine, left leg, pel-vis, and left hand were also obtained. Thepresence of the N540K mutation was veri-fied by restriction enzyme digestions. Halfof our patients carried the N540K mutation.Although similar in phenotype to the pa-tients without the mutation, they showed inaddition relative macrocephaly. The asso-ciation of the unchanged/narrow interpe-dicular distance with the fibula longer thanthe tibia was more common in patients withgene mutation.

Although we did not find a firm correla-tion between genotype and phenotype, in

our study the N540K mutation was most of-ten associated with disproportionate shortstature, macrocephaly, and with radiologi-cal findings of unchanged/narrow interpe-dicular distance and fibula longer thantibia. Am. J. Med. Genet. 75:109–112, 1998.© 1998 Wiley-Liss, Inc.

KEY WORDS: hypochondroplasia; FGFR3mutations; genotype/pheno-type

INTRODUCTION

Hypochondroplasia is an autosomal dominant skel-etal dysplasia characterized by a stocky and muscularphysique. Patients present short-limbed short stature,short and broad hands and feet, genu varum, lumbarlordosis and, exceptionally, mild mental deficiency[Walker et al., 1971; Wynne-Davis et al., 1981].

The clinical and radiological criteria for the diagnosisof hypochondroplasia are controversial. The phenotypesometimes overlaps with that of achondroplasia, whilesevere idiopathic short stature can be erroneously con-sidered as a ‘‘mild form’’ of hypochondroplasia.

The recurrent G380R mutation in the fibroblastgrowth factor receptor 3 (FGFR3) gene was recentlyreported in more than 95% of patients with achondro-plasia [Rousseau et al., 1994; Shiang et al., 1994; Bel-lus et al., 1995a], while the N540K mutation in thesame gene was described in more than 50% of patientswith hypochondroplasia [Bellus et al., 1995b,c; Prinoset al., 1995; Stoilov et al., 1995a; Bonaventure et al.,1996]. As the clinical and radiological diagnosis is notstraightforward, the molecular analysis could be usefulfor re-evaluation of the clinical and radiological find-ings.

Contract grant sponsor: Pharmacia & Upjohn, Italy.*Correspondence to: Chiara Prinster, MD, Department of Pe-

diatrics, H San Raffaele, via Olgettina 60, 20132 Milan, Italy.Received 22 April 1997; Accepted 28 July 1997

American Journal of Medical Genetics 75:109–112 (1998)

© 1998 Wiley-Liss, Inc.

The aims of the present study were 1) to identify thefrequency of the FGFR3 gene mutation in a cohort ofpatients with suspected hypochondroplasia, 2) to de-fine their salient clinical and radiological abnormali-ties, and 3) to verify the contribution of molecular find-ings to the clinical and radiological definition of hypo-chondroplasia.

SELECTION OF PATIENTS

We report on a collaboration involving several Ital-ian departments of pediatrics. Forty-two children withshort stature were enrolled in this study (28 boys and14 girls); median age was 12.1 years (range: 1.56–18.7).The children were either referred for disproportionateshort stature with other physical anomalies suggestiveof hypochondroplasia or they had already been classi-fied as affected by this condition. They were all spo-radic cases except for one daughter and her mother.

Anthropometric measurements included standingheight, sitting height, and head circumference as pa-rameters of body disproportion. All measurementswere repeated three times and the averages were re-corded. Standard deviation scores (SDS) of standingheight and sitting height were calculated. The valuesobtained were plotted on standard curves for lack ofdetailed charts for hypochondroplasia.

Based on the Hall and Spranger [1979] report onhypochondroplasia, we took into consideration only theradiological findings with a frequency greater than80%: 1) narrow or unchanged lumbar interpediculardistance; 2) anteroposterior shortening of lumbar ped-icles (lateral view); 3) posterior scalloping of lumbarvertebral bodies (lateral view); 4) squared shortenedilia; 5) short broad femoral neck; 6) fibula longer thantibia; and 7) mild to moderate brachydactyly. There-fore, standard radiographs were performed of lumbarspine (anteroposterior and lateral view), pelvis (antero-posterior view), left leg (anteroposterior view), and leftwrist and hand. The presence of three of these radio-logical findings was the minimal inclusion require-ment. All metacarpal and phalangeal lengths of the lefthand were measured and length SDS was calculatedfor every osseous segment [Poznanski, 1984].

Radiographs were reviewed separately by G.B. andM.D.M. who examined all the cases. They interpretedthe radiographs independently and only when bothagreed on a finding, this was accepted.

MOLECULAR STUDIES

In the patients considered positive for hypochondro-plasia by radiological criteria, genomic DNA was ex-tracted from whole blood [Miller et al., 1988] and ana-lyzed according to the procedure described by Bellus etal. [1995b]. In addition, we considered the possibility ofdetecting by restriction enzyme digestions both the hy-pochondroplasia mutations (C1620A and C1620G) firstreported by Prinos et al. [1995]. We analyzed the re-striction map of the polymerase chain reaction (PCR)amplified FGFR3 fragment (578 bp) by the DNA-strider software.

Since we found that the C to A mutation abolishes a

restriction site for BspMI, PCR products were digestedwith BspMI. Moreover, to detect the C1620G mutation,which creates an AluI restriction site, we performed adigestion with this endonuclease in all the patients het-erozygous at the restriction analysis with BspMI.

In order to exclude the presence of mutations forachondroplasia (G1138A and G1138C), all subjectswere tested by PCR amplification and restriction en-zyme digestion as described by Shiang et al. [1994].Based on the results of the molecular analysis, we di-vided the patients into two groups depending on thepresence (group 1) or absence (group 2) of the N540Kmutation. Height standard deviation score (HSDS) andtrunk SDS were compared with Student’s t test be-tween the two groups. Values are expressed as mean ±SD. All tests were conducted at the a 4 0.05 level andwere two tailed.

RESULTSGenetic Findings

The genetic analysis was performed in the 18 pa-tients positive by the chosen radiological criteria. Ninepatients carried the N540K mutation of the FGFR3gene (group 1). In seven patients, restriction analysis ofthe 578 bp amplified fragment of the FGFR3 gene re-sulted in the disruption of a site for BspMI, demon-strating the presence of the C1620A mutation (Fig. 1).In two subjects, the creation of a new restriction site forAluI at nucleotide 1620 showed the presence of the C toG substitution (Fig. 1).

Group 2 included all the patients who did not havethe N540K mutation. None of the studied patients pre-sented the achondroplasia G380R mutation.

Auxological Findings

In the patients enrolled in our study, HSDS was −3.0± 0.8 (range: −4.8/−1.6) and mean trunk SDS was −1.2± 1.3 (range: −3.1/0.5). Mean HSDS and trunk SDS ofthe group 1 and 2 patients were −2.7 ± 0.7 and −1.3 ±1.6 and −3.3 ± 0.8 and −2.5 ± 1.3, respectively (Table I).

Fig. 1. Analysis of the FGFR3 amplified product. Left: Detection of theC1620A mutation by restriction digestion: the C to A transversion abol-ishes a site for BspMI. M, DNA molecular weight marker (50–2,000 bpladder). Lane 1: A patient heterozygous for the mutation. Lane 2: Normalcontrol. Right: Detection of the C1620G mutation by restriction digestion:the nucleotide substitution creates a new site for AluI. M, DNA molecularweight marker (50–2,000 bp ladder). Lane 1: A patient heterozygous forthe mutation. Lane 2: Normal control.

110 Prinster et al.

Values of HSDS and trunk SDS did not show signifi-cant differences between the two groups (P 4 0.2 andP 4 0.17, respectively).

Eight patients in group 1 had a head circumference(OFC) > 50 centile and one (an 18-month-old) had anOFC below the third centile. In group 2, OFC was ex-tremely variable (<3 to more than 97 centile) (Fig. 2). Inthe two cases where microcephaly was detected, it wasfamilial (both parents had OFC in the low to normalrange).

Radiological Findings

Radiological findings of the selected patients are re-ported in Table I. The unchanged/narrow interpedicu-lar distance and longer fibula were the most strikingpathological findings in the patients with the gene mu-tation for hypochondroplasia, the frequencies being100% and 89%, respectively. About 50% of the patientswith the N540K mutation had vertebral scalloping,squared shortened ilia, and a short and broad femoralneck.

Group 2 patients showed unchanged or narrowed in-

terpedicular distance and short and broad femoralneck.

The prevalence of brachydactyly evaluated by visualexamination of the radiographs was similar in bothgroups of patients. When metacarpal and phalangealmeasurements of the two groups were compared, nosignificant differences were found. However, thelengths of metacarpals, proximal, and middle phalan-ges were lower than −2 SD in group 1 patients andbetween −1 and −2 SD in group 2 (Fig. 3).

DISCUSSION

Using the commonest radiological criteria for the di-agnosis of hypochondroplasia proposed by Hall andSpranger [1979], we selected 18 patients from among42 children/adolescents with disproportionate shortstature and phenotype known to be compatible withhypochondroplasia. Only 50% of the patients carriedthe N540K mutation, indicating genetic heterogeneityof hypochondroplasia. The results of the genetic analy-sis, made with a technique as reliable as that used byBellus et al. [1995b] but which was simpler andquicker, were in agreement with those of the first stud-

TABLE I. Auxological and Radiological Findings in Hypochondroplasia Patients Carrying theN540K Mutation and in Patients Without the Mutation*

Patients with N540Kgene mutation

Patients withoutN540K gene mutation

Hall and Spranger(%)

Auxological findingsHSDS (M ± SD) −2.7 ± 0.7 −3.3 ± 0.8 —Trunk SDS (M ± SD) −1.3 ± 1.6 −2.5 ± 1.3 —

Radiological findings (%)Lumbar spine

Unchanged/narrow 100 87 80interpedicular distance

Dorsal concavity 44 37 81Pelvis

Squared shortened ilia 57 37 100Femoral neck

Short and broad 44 89 92Fibula ù tibia 89 75 92Brachydactyly 55 55 97

*Our radiological data are compared to those reported by Hall and Spranger [1979] in a cohort of 39 subjects withhypochondroplasia.

Fig. 2. Distribution of head circumference in 18 hypochondroplasia pa-tients. Group 1: Nine patients with the N540K mutation. Group 2: Ninepatients without the N540K mutation.

Fig. 3. Metacarpophalangeal profile pattern in hypochondroplasia pa-tients. s, Patients positive for the N540K mutation; d, patients negativefor the N540K mutation. m, metacarpal; P, proximal phalanx; M, middlephalanx; D, distal phalanx.

Genotype/Phenotype in Hypochondroplasia 111

ies by Bellus et al. [1995b,c] and later work by Bo-naventure et al. [1996].

Particularly, the C1620A transversion represented39% of the hypochondroplasia mutation while theC1620G substitution accounted for 11% of our cases.Although the frequency of the C to A substitution wassimilar to that reported in the French population [Bo-naventure et al., 1996], the C to G substitution fre-quency in the Italian population was half that of theFrench.

The clinical and auxological findings of the patientswere similar, irrespective of the presence or absence ofthe mutation, except for OFC. A relative macrocephalywas, in fact, a common finding in the patients carryingthe gene mutation, suggesting that increased head sizecould be a clinical marker for such patients. This ob-servation partially agrees with the recent report byRousseau et al. [1996], but the different genotype inRousseau’s patients may also account for their differ-ent phenotypes.

Surprisingly, radiological findings did not differenti-ate clearly between patients with or without the mu-tation. However, the association of the unchanged/narrow interpedicular distance with the fibula longerthan the tibia appears to be much more common inpatients with the mutation, but the small number ofcases does not allow a firm conclusion.

It is remarkable that the short and broad femoralneck was a constant finding in patients without themutation, emphasizing the clinical, radiological, andpossibly genetic heterogeneity of hypochondroplasia. Itis also worth underlining that the same frequency ofbrachydactyly, evaluated visually in both groups, washalf that reported by Hall and Spranger [1979]. How-ever, the severity of brachydactyly was clearly evi-denced by measurements of metacarpals, proximal,middle, and distal phalanges of the left hand, indicat-ing that visual demonstration is less reliable.

The difficulty of diagnosing hypochondroplasia isfurther emphasized by comparing our radiological find-ings in patients carrying the gene mutation with thecriteria previously recommended for the diagnosis ofhypochondroplasia. Among the radiological criteriasuggested by Hall and Spranger [1979], at least fourwere only half as frequent in our gene mutation posi-tive patients. We suggest that clinical and radiologicalcriteria for hypochondroplasia be revised, consideringthe results of molecular genetic analysis.

In conclusion, our simple PCR-based technique is re-liable in the identification of the N540K mutation,which is most often associated with disproportionateshort stature, macrocephaly, and with radiologicalfindings of unchanged/narrow interpedicular distanceand fibula longer than tibia. Our data emphasize thedifficulties of a precise diagnosis of hypochondroplasia.Genetic heterogeneity has been postulated in previous

studies [Stoilov et al., 1995b; Hecht et al., 1995; Rous-seau et al., 1996]. However, it remains controversialwhether patients with no demonstrable mutation butwith disproportionate short stature and radiologicalabnormalities of the skeleton should be considered ashaving hypochondroplasia.

ACKNOWLEDGMENTS

We thank Dr. M.C. Proverbio and Mrs. M. Puzzoviofor their technical contribution to this work. We grate-fully acknowledge the financial support of Pharmacia& Upjohn, Italy.

REFERENCESBellus GA, Hefferon TW, Ortiz de Luna RI, Hecht JT, Horton WA,

Machado M, Kaitila I, McIntosh I, Francomano CA (1995a): Achondro-plasia is defined by recurrent G380R mutations of FGFR3. Am J HumGenet 56:368–373.

Bellus GA, McIntosh I, Smith EA, Aylsworth AS, Kaitila I, Horton WA,Greenhaw GA, Hecht JT, Francomano CA (1995b): A recurrent muta-tion in the tyrosine kinase domain of fibroblast growth factor receptor3 causes hypochondroplasia. Nat Genet 10:357–359.

Bellus GA, Szabo JK, McIntosh I, Kaitila I, Aylsworth AS, Hecht JT, Fran-comano CA (1995c): Hypochondroplasia: A second recurrent mutationof fibroblast growth factor receptor 3 (FGFR3) at nucleotide 1620. AmJ Hum Genet 57(Suppl.):A47.

Bonaventure J, Rousseau F, Legeai-Mallet L, Le Merrer M, Munnich A,Maroteaux P (1996): Common mutations in the fibroblast growth factorreceptor 3 (FGFR3) gene account for achondroplasia, hypochondropla-sia, and thanatophoric dwarfism. Am J Med Genet 63:148–154.

Hall BD, Spranger J (1979): Hypochondroplasia: Clinical and radiologicalaspects in 39 cases. Radiology 133:95–100.

Hecht JT, Herrera CA, Greenhaw GA, Francomano CA, Bellus GA, Blan-ton SH (1995): Confirmatory linkage of hypochondroplasia to chromo-some arm 4p [letter]. Am J Med Genet 57:505–506.

Miller SA, Dykes DD, Polesky HF (1988): A simple salting out procedurefor extracting human nucleated cells. Nucleic Acids Res 16:1215.

Poznanski AK (1984): ‘‘The Hand in Radiologic Diagnosis.’’ 2nd ed. Phila-delphia: WB Saunders, pp 31–66.

Prinos P, Costa T, Sommer A, Kilpatrick MW, Tsipouras P (1995): A com-mon FGFR3 gene mutation in hypochondroplasia. Hum Mol Genet 4:2097–2101.

Rousseau F, Bonaventure J, Legeai-Mallet L, Pelet A, Rozet J-M, Maro-teaux P, Le Merrer M, Munnich A (1994): Mutations in the gene en-coding fibroblast growth factor receptor-3 in achondroplasia. Nature371:252–254.

Rousseau F, Bonaventure J, Legeai-Mallet L, Schmidt H, Weissenbach J,Maroteaux P, Munnich A, Le Merrer M (1996): Clinical and geneticheterogeneity of hypochondroplasia. J Med Genet 33:749–752.

Shiang R, Thompson LM, Zhu Y-Z, Church DM, Fielder TJ, Bocian M,Winokur ST, Wasmuth JJ (1994): Mutations in the transmembranedomain of FGFR3 cause the most common genetic form of dwarfism,achondroplasia. Cell 78:335–342.

Stoilov I, Kilpatrick MW, Tsipouras PA (1995a): A common FGFR3 genemutation is present in achondroplasia but not in hypochondroplasia.Am J Med Genet 55:127–133.

Stoilov I, Kilpatrick MW, Tsipouras P (1995b): Possible genetic heteroge-neity in hypochondroplasia. J Med Genet 32:492–493.

Walker BA, Murdoch JL, McKusick VA, Langer LO, Beals RK (1971):Hypochondroplasia. Am J Dis Child 122:95–104.

Wynne-Davis R, Walsh WK, Gormley J (1981): Achondroplasia and hypo-chondroplasia: Clinical variation and spinal stenosis. J Bone Joint Surg[Br] 63B:508–515.

112 Prinster et al.