Fibrodysplasia ossificans progressiva in Spain: epidemiological, clinical, and genetic aspects

Bone 51 (2012) 748–755

Contents lists available at SciVerse ScienceDirect

Bone journal homepage: www.elsevier.com/locate/bone

Original Full Length Article

Fibrodysplasia ossificans progressiva in Spain: epidemiological, clinical, and genetic aspects A. Morales-Piga a,⁎, J. Bachiller-Corral b, M.J. Trujillo-Tiebas c, d, A. Villaverde-Hueso a, d, M.L. Gamir-Gamir b, V. Alonso-Ferreira a, d, M. Vázquez-Díaz b, M. Posada de la Paz a, d, C. Ayuso-García c, d a

Rare Disease Research Institute (Instituto de Investigación de Enfermedades Raras — IIER), Carlos III Institute of Health (Instituto de Salud Carlos III — ISCIII), Madrid, Spain Department of Rheumatology, Ramón y Cajal Hospital, Madrid, Spain Medical Genetics Department, Fundación Jiménez-Díaz, Madrid, Spain d Consortium for Biomedical Research in Rare Diseases (Centro de Investigación Biomédica en Red de Enfermedades Raras — CIBERER), Madrid, Spain b c

a r t i c l e

i n f o

Article history: Received 10 May 2012 Revised 22 June 2012 Accepted 2 July 2012 Available online 13 July 2012 Edited by: J. Aubin Keywords: Fibrodysplasia ossificans progressiva FOP Heterotopic ossification ACVR1 BMP type I receptor

a b s t r a c t We aimed to investigate the epidemiological determinants, clinical features, and genetic pattern of FOP in our country by evaluating the entire population of patients identified according to a combination of methods. To achieve this, 24 individuals were confirmed as FOP cases, 17 of whom were alive at the end of 2011 (point prevalence = 0.36 × 10−6). The gender distribution (male/female ratio = 13/11) and the concurrent range of ages (from 4 to 53 years; mean ± SD: 30.2 ± 13.8) are in agreement with similar reports. Twenty-one (87.5%) had characteristic congenital malformations of the big toe, and short thumbs were found in 65.2% of cases. In addition, other skeletal malformations such us fusion of the posterior elements of the cervical spine (89.0%), knee osteochondromas (71%), scoliosis (54.5%), and short and broad femoral neck (52.6%) were observed. All had developed mature ossicles of heterotopic bone in typical anatomic and temporal patterns, ranging in number from 1 to 17 (9.5 ± 3.9). Age at appearance of first ossifying lesion varied from 3 months to 15 years. Mean age at diagnosis was 7.3 ± 5.1 years and the average delay in reaching the correct diagnosis after the onset of heterotopic ossification was 2.7 years (range = 0–12 years). Biopsy of the pre-osseous lesions was performed in 11 of 20 (55.0%), providing no useful information for the diagnosis of FOP. Seven of 18 (38.9%) reported some hearing loss, and 5 (27.8%) experienced diffuse thinning of the hair or were bald. No patient had relatives with a typical FOP clinical picture. Fourteen of the 16 cases which were genetically investigated displayed the single heterozygous mutation c.617G>A in exon 4 of the ACVR1 gene. One of the two patients who did not present with the canonical ACVR1 mutation showed a heterozygous mutation c.774G>C in exon 5 leading to the substitution of Arginine 258 with a serine. The other patient had a heterozygous c.774G>T substitution in exon 5 leading to the same amino acid change (p.Arg258Ser). These two patients had only nonspecific abnormalities of the great toe, lacked the typical anatomic and developmental pattern of heterotopic ossification, and shared a trend toward uncommon clinical features. These results provide new insight on the epidemiological and clinical traits of FOP, reinforcing the notion of its worldwide homogeneity. The molecular characterization of ACVR1 sequence variation will contribute to the understanding of the genetic profile of this devastating disease in different geographical areas. © 2012 Elsevier Inc. All rights reserved.

Introduction Fibrodysplasia ossificans progressiva (FOP) is the most severe and disabling disorder of extraskeletal ossification in humans [1–3]. Its main characteristics are typical congenital skeletal malformations ⁎ Corresponding author at: Instituto de Salud Carlos III, Instituto de Investigación de Enfermedades Raras, Monforte de Lemos, 5, 28029, Madrid, Spain. Fax: +34 1 387 78 95. E-mail addresses: [email protected] (A. Morales-Piga), [email protected] (J. Bachiller-Corral), [email protected] (M.J. Trujillo-Tiebas), [email protected] (A. Villaverde-Hueso), [email protected] (M.L. Gamir-Gamir), [email protected] (V. Alonso-Ferreira), [email protected] (M. Vázquez-Díaz), [email protected] (M. Posada de la Paz), [email protected] (C. Ayuso-García). 8756-3282/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2012.07.002

and progressive heterotopic ossification following specific anatomic patterns. Most cases arise as a result of a spontaneous new mutation and fewer than ten small families are known to have the disease [4]. Its worldwide prevalence is approximately one in two million individuals [1–3], and there appears to be no ethnic, racial, gender, or geographic predisposition [4–6]. Diagnostic errors are frequent in FOP [2,3]. Many patients are misdiagnosed before heterotopic ossification develops, causing them to undergo both unnecessary and even harmful diagnostic and therapeutic procedures. However, an accurate diagnosis can be made early in life based on the presence of tumor-like swellings and characteristic malformed great toes [7]. The recent discovery of a recurrent mutation in the gene encoding the activin A type I receptor (ACVR1), considered

A. Morales-Piga et al. / Bone 51 (2012) 748–755

to be the cause of virtually all classically occurring inherited and sporadic cases of FOP [8], allows for reliable confirmatory diagnosis before ectopic ossification appears, thus avoiding inappropriate medical procedures [3]. Since the first accounts of FOP in the early 1980s [9,10], a number of individual cases have been reported in Spain [11–20]. While most of these reports provide a fairly thorough description of isolated cases, overall reviews are lacking. As a result, the epidemiological facts of the disease in our area remain unknown. Moreover, to our knowledge, most of our neighboring countries also lack data from systematic purpose-designed surveys. Given this scarcity of evidence, it is currently impossible to make comparisons between geographical regions and, thus, the differences in prevalence or in clinical characteristics are unable to be traced. The absence of an effective treatment for this devastating disease requires further investigation into these questions in order to explore effective preventive and therapeutic measures. Our work sought to investigate the epidemiological determinants, clinical features, and genetic pattern of FOP in Spain by evaluating the entire population of patients identified by a national survey. Material and methods Diagnosis of FOP was based on the concomitant presence of [2]: 1) Characteristic congenital malformations of the great toe (hallux valgus, malformed first metatarsal, and/or monophalangism) and, 2) Ribbons, sheets, or plates of heterotopic ossification with radiologic appearance of lamellar mature bone, which develops inside of soft connective tissues and follows specific anatomic and temporal patterns. Complete ascertainment of FOP in Spain was attempted using several methods: 1) Individual cases seen in a referral unit specializing in bone diseases and recorded by two of us (A M-P and J B-C) over different periods (1980 to 2002 and 2002 to present). 2) A survey of associations of patients with genetic bone disorders, particularly AEFOP (Spanish Association of FOP). Once the diagnosis was confirmed, the case information was updated in order to meet the requirements of our study. 3) A national survey announced in the official journals and computer networks of scientific organizations most likely to be involved in FOP patient care: the Spanish Society of Rheumatology, the Spanish

749

Society of Family and Community Medicine, and the Spanish Society of Human Genetics. 4) In addition, we have reviewed the papers published in our country in which well‐documented cases of FOP are depicted. Whenever possible, we have contacted the corresponding author in order to either receive the materials needed for a correct assessment in accordance with our requirements or, alternatively, we have asked that they refer us to the patient himself to update the study. Based on fulfillment of any of the above-mentioned methods, a total of 24 individuals were identified and confirmed as being FOP cases (Fig. 1). Five were personally followed-up in our hospital, either from the beginning or for the most part of the course of their disease. Complete and updated information from a second set of 7 patients studied in different hospitals across the country was sent to us by the Spanish FOP Association (AEFOP). After reviewing their medical records, we can conclude that all were well documented FOP cases. As a result of the national survey carried out with the assistance of scientific organizations we learned of a relatively large number of presumptive patients. However, after ruling out duplicates and inconsistencies, just 4 of them were confirmed as being true “new FOP cases” that had not yet been identified by other means. In three of these four we gained full access to their medical records and the appropriate information was recorded. Finally, by carrying out a systematic bibliographical search, we recovered a total of 21 articles published in Spain between 1980 and 2010 depicting 23 cases in which the term FOP appears as the diagnosis. Among them were the two initial cases reported by our group in 1982 [10] and several other cases already identified in the present study by other means [11–14]. After refining the search and by excluding the doubtful or incompletely documented reports, we were able to confirm 8 additional (previously unknown) FOP cases [9,15–20]. Three of them were updated for us based on information provided by the authors (or current MD in charge). The remaining five published cases were lost to follow-up and, consequently, we used their information only for some isolated descriptive purposes. Independent of the procedure which led to the discovery of a potential FOP case, all the relevant records and radiographs were reviewed personally by two of us (A M-P and J B-C). After confirming the diagnosis, we sent out a questionnaire requesting additional information on personal – medical as well as social – and familial issues. Where possible, face-to-face visits or telephone calls were arranged to update and narrow down the information and also to interview relatives.

Fig. 1. Number of cases (inside circles) according to procedure of recruitment and quality of the resulting information.

750

A. Morales-Piga et al. / Bone 51 (2012) 748–755

This study was approved by the Ethical Committee of the Rare Disease Research Institute of the Carlos III Institute of Health (Madrid, Spain). All subjects were asked to sign an informed consent form and it was proposed that each donate a biological sample for genetic study. In all of the 16 FOP probands in whom a sample of blood or saliva was obtained, bi-directional automatic sequencing was performed in order to screen the ACVR1 gene for mutations. Saliva (n = 1) or peripheral blood (n = 15) samples were collected in ethylenediaminetetraacetic acid (EDTA) tubes and genomic DNA was extracted using an automated extractor (model BioRobotEZ1; Qiagen, Hilden, Germany) following manufacturer instructions. First, the exon 4 of the ACVR1 gene (including intron–exon junctions) was amplified by PCR with primers as previously described [8]. These fragments were electrophoresed in a 3% agarose gel and purified using a DNA extraction kit (E.Z.N.A. Cycle Pure Ki; Omega Bio-Tek). Sequencing reactions were performed using the 4 dye terminator cycle sequencing ready reaction kit (BigDye DNA Sequencing Kit v. 1.1, Applied Biosystems, Foster City, CA). Sequence products were purified through fine columns (Sephadex G-501, Princetown

Separations, Adelphia, NJ) and analyzed in an ABI Prism 3130 (Sequencing Analysis v. 5.2; Applied Biosystems). In the cases which did not display the canonical c.617G>A mutation (in exon 4) we analyzed the exons 3 and 5 of the ACVR1 gene, following the same procedures. The sequences and annealing temperatures used are available from the authors on request. The significant differences (p b 0.05) were determined using the two-tailed Student's t test (or non‐parametric test when appropriate) and the chi-square test with Yates's correction. Results A total of 24 individuals were identified and confirmed as FOP cases (Fig. 1). Of these, three (two women and one man) had died before the study was undertaken, and four could not be traced. The age and causes of death were: pneumonia at the age of 44 (case # 9; Table 1), pneumonia and heart failure at 46 (case # 10), and pneumonia and septic shock at an undetermined age (case # 24). The remaining 17 were alive in Spain at the end of 2011 (total estimated population:

Table 1 Main clinical characteristics of 24 Spanish fibrodysplasia ossificans progressiva patients. Case

Gender

Age (yr.)

Skeletal malformations

Ectopic ossification

Great-toe malformation

Others

Age (yr.)

Onset site

N° lesions

Short thumbs/exostosis (hands) Cervical spine fusions Short thumbs/clinodactyly Cervical spine fusions Broad femoral necks/knee osteochondroma Short thumbs/clinodactyly Knee osteochondroma Short thumbs/cervical spine fusions Broad femoral necks/knee osteochondroma Clinodactily/cervical spine fusions Broad femoral necks Cervical spine fusions/broad femoral necks Double patella/exostoses Clinodactily (No complete Rx study) Short thumbs/cervical spine fusions Clinodactily/exostosis/scoliosis/synostosis Short thumbs/cervical spine fusions Broad femoral necks/knee osteochondroma Short thumbs/cervical spine fusions Knee osteochondroma Short thumbs/cervical spine fusions Broad femoral necks/knee osteochondroma Clinodactily/cervical spine fusions Knee osteochondroma Short thumbs/cervical spine fusions Knee osteochondroma/scoliosis Cervical spine fusions/synostosis Knee osteochondroma/exostoses (hip) Short thumbs/Clinodactyly/cervical spine fusions Broad femoral necks/knee osteochondroma Clinodactily/cervical spine fusions Scoliosis Knee osteochondroma (No complete Rx study) Clinodactily/knee osteochondroma/scoliosis Short thumbs/clinodactyly/cervical spine fusions Broad femoral necks/exostosis/scoliosis Short thumbs/clinodactyly Broad femoral necks (No complete Rx study) Short thumbs/cervical spine fusions/scoliosis Broad femoral necks Short thumbs/exostosis Short thumbs/clinodactyly/scoliosis

11

Shoulder

6

Mild

1

5

Neck

17

Mild

3

0.4

Neck

12

No

3

Neck

13

No

0.4

Neck

10

No

6

Hip

10

No

2.5

Neck

–

–

0.8

Neck

8

No

–

Unclear

14

Light

–

Unclear

15

Light

1.5

Neck

7

No

2

Paraspinal

9

Severe

15

Hip

4

No

11

Hip

1

No

0.8

Neck

8

Light

14

Scapula

10

No

5

Unclear

7

No

6 0.2

Neck Neck

12 9

No Light

0.1 4

2

Knee

–

–

3

2.5 1.5

Neck Neck

– –

– –

2.5 –

1.1 9

Paraspinal

– –

– –

0.9 –

1

F

26

Typical

2

M

41

Typical

3

M

4

Typical

4

M

23

Typical

5

F

21

Typical

6

F

42

Atypical

7

M

10

Typical

8

F

33

Typical

9

F

44

a

10

F

46

a

11

F

13

Typical

12

F

40

Typical

13

M

29

Atypical

14

M

13

Typical

15

F

12

Typical

16

F

53

Atypical

17

M

52

Typical

18 19

M M

29 26

Typical Typical

20

F

37

Typical

21 22

M M

32 37

Typical Typical

23 24

M M

43 ¿? a

Typical Typical

See text for more details. a Age at death.

Typical Typical

Deaf

Other features

Cutaneous angioma

Diagnostic delay (yr.)

0.4 0.5

Persistence of primary teeth Bald Abnormal dentition

1.7 10 0.4

Short height

1.1 –

Bald Leiomioma Persistence of primary teeth/hypertrichosis Bald Abnormal dentition Bald

– 0.5 12 1 2

Hypertrichosis

5.2 2

Low IQ/bald

2

A. Morales-Piga et al. / Bone 51 (2012) 748–755

47,150,819 [21]), yielding a point prevalence of 0.36 × 10−6. As an indication of risk, we have divided the estimated number of cases from our cohort who were born between the years 1970–2009 (n= 17) by the number of newborns in Spain registered by the Spanish Institute of Statistics for the same period (19,705,970) [21], which is equal to one case out of 1,150,000 live births. The risk increases when only children of mothers older than 40 years are considered (1/240,000) as well as when the fathers are above this age (1/345,000). When both parents are older than 40 years, the estimation of risk rises to 1/130,000. The clinical characteristics of the 24 probands are depicted in Table 1. The male/female ratio was 13/11, and the concurrent age ranged from 4 to 53 years (mean ± SD: 30.2 ± 13.8). Twenty-one (87.5%) have characteristic congenital malformations of the big toe, whereas the remaining 3 (cases # 6, 13, and 16) show only minor atypical abnormalities in this location (Figs. 2a and 3a). Short thumbs were found in 15 of the 23 (65.2%) for whom the information was available and clinodactyly of the fingers was present in 12 of 23 (52.2%). Additionally, proven cases showed other skeletal malformations such us fusion of the posterior elements of the cervical spine (16 of 18 available cases; 89.0%), knee osteochondromas (12 of 17; 71%), scoliosis (12 of 22; 54.5%), and short and broad femoral neck (10 of 19; 52.6%). All probands had developed mature ossicles of heterotopic bone in typical anatomic and temporal patterns, ranging in number from 1 to 17 (mean±SD: 9.5±3.9). None of the patients were diagnosed at birth. The ages at appearance of first ossifying lesion in 22 cases in whom such a finding was recorded show a wide variability (from 3 months to 15 years), with the majority (15 of 22) experiencing the onset of slumps prior to the age of five. The mean age (±SD) at diagnosis was 7.3±5.1 years and the average delay in reaching the correct diagnosis after the onset of heterotopic ossification was 2.7 years, with a range of 0 to 12 years. Eight of 18 (44.4%) patients reported muscular trauma prior to the time the lumps became apparent. The site of onset was usually the neck (12; 52.2%) followed by hip (3; 13.0%), the shoulder/ scapula (2; 8.7%), and the dorsal–paraspinal region (2; 8.7%). Six of the 17 patients (35.3%) who were interviewed reported that they had noted pain upon the appearance of new lumps. With the passage of time, certain areas were especially prone to ossification, specifically the

Fig. 2. Plain radiographs from patient #6 (lacking the canonical mutation) showing: a. Exostoses at the distal end of first metatarsal and symphalangism of the great toe, with no characteristic hallux valgus deformity. b. Extensive heteroptopic ossification at the rear distal femur with characteristic endochondral appearance and creation of true joint structures. It is also worth noting the presence of double patella.

751

Fig. 3. Plain radiographs from patient #13 (lacking the canonical mutation) showing: a. First metatarsal exostoses (bilateral) with no shortening or characteristic hallux valgus deformity. b. Plates of heteroptopic ossification around the hip and (less prominent) at lumbo-sacral area.

connective tissue of the paraspinal muscles, the muscles of the limb girdle, and the muscles of mastication. Eleven of 20 (55.0%) had undergone a biopsy of the pre-osseous lesions; only in three of these cases was this procedure able to help rule out other diseases, but no additional information was obtained from any of these procedures which led to the diagnosis of FOP. Nine of 20 patients (45.0%) had been subjected to one or more surgical procedures (other than biopsy) because of the disease. Seven of the 18 (38.9%) reported some hearing loss, and 5 (27.8%) experienced diffuse thinning of the hair or were bald; 3 such patients were female (# 6, 10, and 12). Mental development was sub-normal in one patient. Seven of 18 (38.9%) stated that they were able to undertake near-normal physical activity, 8 (44.4%) required the use of crutches, a walker, or a wheelchair, and 3 (16.7%) were unable to walk at all. None of the 24 probands had reproduced. We did not attempt to survey in detail the results of drug therapy. However, all but 3 patients in our cohort (83.3%) had taken at least one bisphosphonate at some point during the course of the disease. Four patients considered this therapy somewhat helpful. Eleven of 18 (61.1%) patients received early, short-course treatment with oral corticosteroids usually in the presence of new flare-ups involving major peripheral joints, and systematically in the management of submandibular swelling. In all, 8 of these patients thought that they experienced some general (unspecific) improvement after taking corticosteroids. Parental age at the time of birth (rounded to the nearest year) was available for 18 of 24 FOP probands; the mean values (± SD) were 34.4 (±8.5) for fathers and 30.3 (± 6.7) for mothers. A comparison of these data with mean values for parental ages at the time of birth in Spain calculated in 1980 (average year of birth for our cohort) showed a higher statistically significant age of fatherhood than in the general population (n= 557,378; mean± SD: 30.1± 6.2; p = 0.03). Regarding exposure to environmental factors, in the years immediately before birth one father had been repeatedly exposed to chemicals and

752

A. Morales-Piga et al. / Bone 51 (2012) 748–755

another six fathers had been exposed to several chemical and/or environmental agents, though to a lesser extent. Three mothers had had some exposure to chemical and/or environmental agents with the potential to induce mutations. Five out of 19 pregnancies (26.3%) for which data could be obtained were considered anomalous. And 6 of 19 cases (31.6%) reported some type of disturbance in the delivery of the child who later developed FOP. No patient had a relative with the typical FOP clinical picture. However, 6 of 18 FOP cases (33.3%) have one or various first-degree relatives with some anomaly of the feet (in some cases early-onset hallux valgus) or other minor skeletal congenital defects (like clinodactyly or polydactyly), often resembling those seen in the correspondent index case. Congenital bicuspid aortic valve, Kartagener syndrome, and Down syndrome (with associated cardiac malformation) were other genetic diseases recorded in first-degree relatives of the probands. None of the parents were consanguineous, though the grandparents of one of the patients (# 6) were first cousins. The screening for mutations in the ACVR1 gene carried out in the 16 cases in which a sample was obtained (15 blood and one saliva) revealed 14 such cases with the canonical c.617G>A mutation in

exon 4 (Fig. 4). In one (# 6) of the two patients who did not present with this mutation we found a heterozygous mutation c.774G>C in exon 5, leading to the change of Arginine 258 with a serine (Fig. 4). In the other case (# 13) which was negative for the canonical FOP mutation further analysis led to the identification of a heterozygous c.774G>T substitution in exon 5 leading to the same amino acid change (p.Arg258Ser) in the kinase domain of the protein (Fig. 4). Table 2 shows the correspondence between age of onset and gender, clinical features, evolutionary hallmarks, and unusual traits of FOP, and the presence/absence of the ACVR 1 mutation in the 16 patients genetically investigated. According to the proposals of Kaplan et al. [22], 12 of the 14 patients with the canonical c.617G>A mutation met all the requirements to be labeled as “classic” forms of FOP by fulfilling the previously defined features of: 1) typical malformation of the great toe, and 2) characteristic FOP-type heterotopic ossification: early-onset (except patient #1) and progressive characteristic temporal and anatomic pattern. Patient number 14 showed the typical malformation of the great toe but, as regards heterotopic calcification, the disease was not clinically apparent until the age of 11 and, although it is as of yet

Fig. 4. Partial electropherograms of exons 4 and 5 of the ACVR1 gene showing the mutations.

A. Morales-Piga et al. / Bone 51 (2012) 748–755

753

Table 2 Patterns of association between phenotype expression and the presence of ACVR 1 mutation in 16 FOP patients. ID case

1

2

3

4

5

6

8

11

12

13

14

15

16

17

18

19

F 11

M 5

M 0.4

M 3

F –

F 6

F 0.8

F 1.5

F 2

M 15

M 11

F 0.8

F 14

M 5

M 6

M 0.2

X

X

X

X

X

X

X

X

X

X

X

X

X

X X ? ? X

X X X X X

X

X X

X X X X

X X

? ? X

? ? X

X X

X

X X X X

0.40 X X

0.47 X X

3.33 X X

Clinical features Gender [M. F] Age of heterotopic ossification (HO) onset [years] Classic/defining FOP features – Typical congenital malformations of great toes – Other common FOP features: a) Congenital malformations of thumbs b) Orthotopic fusions of cervical spine c) Knee osteochondromas d) Short/broad femoral necks e) Conductive hearing impairment Evolutionary hallmarks – No. of mature HO lesions/years of evolution – Progressive HO in characteristic anatomic pattern – Progressive immobility Atypical clinical FOP traits – Bald – Cognitive impairment – Persistence of primary teeth/abnormal dentition ACVR 1 canonical mutation [c.617G>A]

X

X

X

X

X X X

X X

X 0.65 X X

0.51 X X

0.27

0.24 X

X

0.60 X X

X

0.23 X X

0.28

X

X

0.50

X X X X X 0.71 X X

X

X X 0.25 X X

0.14 X X

0.50 X

0.34 X X

Y

Y

X Y

Y

Y

Y

Y

X *

Y

X Y

X Y

**

Y

Y

Y

Y

?: no appropriate Rx available for evaluating this. *[c.774G>C] [27] (Bocciardi et al.). **[c.774G>T] [45] (Ratbi et al.).

too soon to reach a definite conclusion, the course of the disease seems distinctly “benign”, lacking the habitual temporal and topographical pattern. The remaining patient (# 16) who carries the canonical mutation shows minor atypical abnormalities of the great toe, scant associated skeletal malformations (only fusion of the cervical spine) and a relatively mild evolution of heterotopic ossification, with a late age of onset (14 years). The only two patients (# 6 and 13) who failed to demonstrate the canonical c.617G>A mutation display a late age of onset (six and 15 years respectively) and a pattern of heterotopic ossification which is somewhat different, with only minor atypical abnormalities of the great toe (Figs. 2a and 3a). In addition these two cases had a remarkably similar set of peculiar clinical features such as baldness (both cases vs. only one among the 13 with canonical ACVR1 mutation) and persistence of primary teeth (patient # 6). Discussion Taking into account the extreme worldwide rarity of FOP [1–3], our cohort represents a relatively large series of proven cases. For an estimated population of 47,150,819 inhabitants in Spain in 2010 [21], and considering as valid a prevalence of one case for every two million individuals [1–3], the expected number of FOP cases would be about 23. Therefore, the 17 patients who were alive by the end of 2011 (point prevalence = 0.36 × 10 −6) and the 24 included in the whole cohort represent a high proportion of FOP morbidity in our country. This retrospective cross-sectional study of Spanish FOP population follows a similar approach to other studies performed in the USA [23–25] and European countries [26,27],—as well as in other distant geographical areas [28]. Our method of ascertainment, based on a combination of procedures and conceived as a national search (rather than focussing on the patients identified in hospitals), makes this a highly representative sample of unselected cases. One of the goals of this work was to investigate the clinical picture of FOP in our region. Although the number of cases is limited by the extreme rarity of the disease, certain generalizations seem nonetheless possible. The gender distribution (nearly equal sex distribution) and the range of ages – from 4 to 53, with a mean of 30 years – are in agreement with similar reports [23,26]. Although the clinical features of FOP are, at the same time, striking and very characteristic, in our investigation as well as in previous reports [23,29] there is a long delay between the first flare-up of pre-osseous swelling (an

event which is often not easy to time) and the definitive diagnosis. Clearly, this retardation is linked to a failure to appreciate the significance of the abnormal big toes. Shortened big toes are not always identified during the standard (usual) examination of newborns, and even when this anomaly is identified, it is often erroneously understood as an isolated congenital hallux valgus [30]. On the contrary, typical great-toe malformation – present in virtually all FOP patients (87.5% in our series) – is suggestive of a highly regulated developmental expression of phenotypic anomalies in embryogenesis [31]. Therefore, congenital hallux valgus is strongly indicative of FOP and its presence must always prompt assessment of the potential existence of the condition as a first step [25,30]. In accordance with previous reports, skeletal malformations other than those of the big toes were seen in our patients at a variety of sites. Thus, the frequency of fusion of the posterior elements of the cervical spine (89%), short and broad femoral neck (52%), and scoliosis (54%) was similar to the values reported in other comparable series [23–26]. Regarding knee osteochondromas, a poorly recognized manifestation in initial reports, we have found this anomaly in 12 of 17 cases (71%). This result is more in line with a recent work which considers tibial osteochondromas to be a common phenotypic feature present in up to 90% of 96 new FOP cases [32]. However, the actual frequency and significance of this lesion have been subsequently called into question [33], and this issue requires further investigation. Regarding heterotopic bone formation, as noted both in the present study and the existing literature [23–26], onset of this condition usually occurs in early childhood in the neck or the dorsal paraspinal muscles, and the number and topographic distribution of mature lesions that we have found is similar to those depicted in other reports. Diagnostic errors are common in FOP [2,3,34] and, as a consequence, many patients suffer unnecessary diagnostic and therapeutic procedures which can exert a detrimental effect on the course of the disease [7,34]. In our series as well as in previous reports [23–26], biopsy is neither helpful nor necessary to establish the diagnosis. Other surgical procedures, such as excision of ectopic bone or attempts to increase the mobility of a joint by orthopedic corrections, are often useless and worsen the progression of the disease, in some cases even causing serious harm. As a rule, therefore, these procedures must be strongly discouraged. We have observed a constellation of symptoms and pathological traits which may occasionally be seen in association with FOP.

754

A. Morales-Piga et al. / Bone 51 (2012) 748–755

Among these often neglected manifestations, the most frequent are: hearing loss, baldness of the scalp, and persistence of primary teeth. Deafness has been noted in seven of 18 patients (39%), a proportion which is higher than that communicated by Rogers and Geho (12% of 42 cases) [23] and similar to that observed by Connor and Evans (24% of 44 cases) [26]. Although this relationship has been overlooked for years, hearing impairment – whose onset is usually in childhood and may be slowly progressive – is now considered to be a common feature of FOP with an incidence of up to 50% [2]. On the other hand, five patients (28%) in this cohort – three of whom are female – showed diffuse thinning of the hair or were bald. Diffuse baldness of the scalp was present in 24% of the patients investigated by Connor and Evans and, like ours, most of them were women whose hair had become thin in middle age [26]. Whether or not this loss of hair is a primary feature of FOP or secondary to a nutritional deficiency caused by lockjaw remains to be established. Our data on age of fatherhood are consistent with a paternal age effect, a finding initially communicated by Rogers and Geho [23] and shortly thereafter confirmed by Connor and Evans [2]. In addition, we have recorded a relatively high number of environmental agents (especially chemical products) in fathers (and, to a lesser extent, in mothers). Although the validity of these observations is limited due to the paucity of data and the lack of dose-effect estimation, it seems wise to consider these exposures (capable of inducing mutations) in FOP causality. All but two cases reported in this Spanish series display the single de novo heterozygous mutation (617G>A; R206H). This same mutation has been reported in sporadically affected individuals with classic FOP features observed in many other Caucasian [35] and Asian populations like those of China [36], Japan [37], and Korea [38]. This represents a mutation frequency of the ACVR1 gene that is similar to the one previously reported by Connor and Evans [26] (1.8 × 10 −6 mutation per gene per generation), which is similar to those seen in Aniridia [39] or Marfan syndrome [40]. However, in FOP cases the mutation nearly always occurs within the same nucleotide, as opposed to the PAX6 or FBN gene, where de novo mutations arise anywhere in the coding or regulatory elements of these genes. For de novo point mutations, a paternal-age effect must be considered as a predisposing factor, as has been observed in the present series. This particular c.617G substitution constitutes a mutational hotspot since G → A transitions located in a CpG dinucleotide together with C → T, occur in approximately a ten-fold-higher proportion than would have been expected by chance alone [41,42]. So the recurrence of this de novo mutation can be explained by the advanced paternal age and the existence of a hotspot mutational site. Additionally, a selective advantage for the carrier sperms has been reported for some recurrent de novo mutations in different instances such as constitutive activation of FGFR2 [43,44] and mutations of RET proto-oncogene [42]. In our series, no proband had a relative with a complete picture of FOP, and it can therefore be concluded that all were fresh mutations. However, the significance of skeletal abnormalities (including several relatives with early-onset hallux valgus) observed in one-third of their firstdegree relatives as well as the existence of some background of genetically conditioned diseases in some cases requires further explanation. We attempted to identify possible patterns of association between phenotype expression and the presence of the ACVR 1 mutation (Table 2). As a result, it can be stated that 12 of the 14 patients who have the canonical c.617G>A mutation fit the “classic” phenotype described by Kaplan et al. [22]. The other two patients with the canonical ACVR 1 mutation, one of whom lacks the typical malformation of the great toe, show some distinct characteristics: both display a late onset of an unusually mild heterotopic calcification process. As for the two patients in whom the canonical ACVR1 mutation was absent – having only nonspecific abnormalities of the great toe (Figs. 2a and 3a) – their onset of heterotopic ossification tended to be late (Table 2) and were topographically atypical (Figs. 2b and 3b),

both commencing in the hip. Moreover, the number of mature ossifying lesions which they have (one and two, respectively) is well below the average number for the cohort and do not conform to the typical anatomic developmental pattern. Lastly, these two cases share a trend toward development of atypical clinical features such as baldness (both cases vs. only one among the 13 with canonical ACVR1 mutation) and persistence of primary teeth. These two patients who were negative for the common c.617G>A mutation themselves carry other mutations in the ACVR1 gene which have recently been described in several FOP cases from Italy [27], and Morocco [45]. Our case #6 carries the same heterozygous mutation c.774G>C described by Bocciardi et al. [27] in two of the 17 Italian FOP cases from their series. Very interestingly, one of these Italian patients with the newly identified R258S substitution is described as “showing a FOP variant phenotype” with the conspicuous absence of characteristic great-toe malformation, a peculiarity shared with case #6 in our series (Fig. 2a). The Moroccan patient who carries the c.774G> T substitution is not described in great detail or with exactness, but it could be assumed that he also has an unusual clinical pattern with a late age of onset (8 years) and just “short first metatarsal” (without mention of congenital “hallux valgus”) [45]. Again, atypical similar characteristics are present in patient #13 in our series (Table 2), who carries the same uncommon FOP mutation. “These c.774G>T mutations, as well as others like c.1067G>A, which are also reported in phenotypic benign variants of FOP with unusual clinical traits [46], occur in the kinase domain of the protein. The occurrence of a GS domain mutation (c.587T>C) in another case with a slower, mild clinical course [47] shows that the disease course is not related to a domain-specific effect, but rather depends upon the degree of structural perturbation within each mutant protein. Interaction of the genetic setting with inflammatory and immunological factors, mesenchymal stem cells and hypoxic microenvironment has each been identified as emerging determinant clues in the skeletal metamorphosis which FOP means [48].” In conclusion, a better understanding of epidemiological and clinical traits and the molecular characterization of ACVR1 sequence variation in a cohort highly representative of a population settled in a previously ignored geographical area will contribute to a fuller understanding of the genetic variation of the disease. Ultimately this sharper perspective may lead to the identification of molecular targets for the development therapeutic agents whose aim is to control the course of this devastating disease.

Acknowledgments We thank the Spanish FOP Association (AEFOP) – particularly to Patricia Marín –, the patients, and their families for their participation in this study. Also, we wish to thank the consultants throughout Spain who have collaborated in this investigation, especially Drs. GarcíaCallejo F.J. (“Hospital Clínico Universitario”, Valencia), González Herranz P. (“Hospital materno-infantil Teresa Herrera, La Coruña”), Olmedo J (“Hospital Clínico”, Madrid), Solsona B. (“Hospital General Universitario”, Valencia), Solano P. (Jerez de la Frontera), MedranoSan Idelfonso M. (“Hospital Miguel Servet”, Zaragoza), Galán-Gómez E. (“Hospital materno-infantil”, Badajoz), Casado C. (Madrid), PerezGómez M. (Hospital Principe de Asturias, Alcala de Henares), Merino R. (Hospital La Paz, Madrid) and Salmoral M.A. (Hospital Reina Sofía, Córdoba). We also wish to thank Camilo Vélez Monsalve (T.E.L. Genetics Department, Fundación Jiménez-Díaz) and Ana Cristina Muñoz Boyero for their technical assistance in mutational studies and we are grateful to Oliver Shaw for his editorial assistance.

References [1] Connor JM, Evans DAP. Genetic aspects of fibrodysplasia ossificans progressive. J Med Genet 1982;19:35–9.

A. Morales-Piga et al. / Bone 51 (2012) 748–755 [2] FOP Rx Guidelines: The International Clinical Consortium on FOP. The medical management of fibrodysplasia ossificans progressiva: current treatment considerations. Clin Proc Int Clin Consort FOP 2011;4:1–95. [3] Kaplan FS, Le Merrer M, Glaser DL, Pignolo RJ, Goldsby RE, Kitterman JA, et al. Fibrodysplasia ossificans progressiva. Best Pract Res Clin Rheumatol 2008;22:191–205. [4] Shore EM, Feldman GJ, Xu M, Kaplan FS. The genetics of fibrodysplasia ossificans progressiva. Clin Rev Bone Miner Metab 2005;3:201–4. [5] Kaplan FS, Shore EM, Connor JM. Fibrodysplasia ossificans progressiva (FOP). In: Royce PM, Steinmann B, editors. Connective tissue and its heritable disorders: molecular, genetic, and medical aspects. 2nd ed. New York: Wiley-Liss, John Wiley & Sons, Inc.; 2002. p. 827–40. [6] Kaplan FS, Glaser DL, Shore EM. Fibrodysplasia (myositis) ossificans progressiva. In: Favus MJ, editor. Primer on the metabolic bone diseases and disorders of mineral metabolism. 6th edn. Washington, DC: The American Society for Bone and Mineral Research; 2006. p. 450–3. [7] Kitterman JA, Kantanie S, Rocke DM, Kaplan FS. Iatrogenic harm caused by diagnostic errors in fibrodysplasia ossificans progressiva. Pediatrics 2005;116:e654–61. [8] Shore EM, Xu M, Feldman GJ, Fenstermacher DA, Cho TJ, Choi IH, et al. A recurrent mutation in the BMP type 1 receptor ACVR1 causes inherited and sporadic fibrodysplasia ossificans progressiva. Nat Genet 2006;38:525–7. [9] Bayes R, Narbona E, Muñoz A, Galdo G, Molina JA. Fibrodisplasia osificante progresiva. Enfermedad de Muchmeyer. Arch Pediatr 1980;31:527–36. [10] Olmedo J, Morales Piga A, García JA, Barrios R, Zea A, Beltrán J. Fibrodisplasia osificante progresiva. Med Clin (Barc) 1984;83:336–8. [11] Del Pino Ortiz, Gómez de Membrillera G, Laguía M. Fibrodisplasia osificante progresiva: aportación de un caso y resultado del tratamiento con difosfonatos. Rev Esp Cir Osteoartic 1987;22:117–23. [12] Fernández Toral J, Martínez Noval MJ, Queiro Silva R, Ballina García FJ, Barreiro Daviña J, Cobo Ruisánchez A. Miositis o fibrodisplasia osificante progresiva idiopática (Enfermedad de Munchmeyer). Act Ped Esp 1994;52:485–90. [13] García Callejo FJ, Morant Ventura A, Orts Alborch MH, Blay Galaud L, Marco Algarra J. Lesiones en cabeza y cuello por Fibrodisplasia osificante progresiva (Enfermedad de Munchmeyer). Acta Otorrinolaringol Esp 2000;51:646–54. [14] Duplá Arenaz M, Torres Claveras S, Fernández Vallejo B, Suso Fernández M, Pastor Mourón I. Bultoma cérvico-escapular. An Pediatr (Barc) 2006;6:511–2. [15] Sánchez E, Maturana I, Delgado A. Miositis osificante progresiva. Aportación de un caso y resultados del tratamiento con etidronato disódico. An Esp Pediatr 1984;21:597–601. [16] Vicens A, Arandes JM, Piulachs J, Suñol J, Rives A, Martinez F. Fibrodisplasia osificante progresiva generalizada. Presentación de un caso y revisión de la literatura. Barc Quir 1987;30:251–63. [17] Martínez Assucena MA, Mañez Añón I, Iñigo Huarte VE, Núñez Cornejo Piquer C. Miositis osificante progresiva. Estudio evolutivo de un caso. Med Rehabil 1989;11:67–70. [18] Sánchez Castro JM, Castaño Sobrón A, Facal Varela M, Trobajo de as Matas JE. Fibrodisplasia osificante progresiva. Rev Ortop Traumatol 1995;39:51–3. [19] Pérez-Seoane Cuenca B, Merino Muñoz R, de José Gómez MI, García-Consuegra Molina J. Fibrodisplasia osificante progresiva: aportación de 2 casos. An Pediatr (Barc) 2006;64:173–86. [20] Salmoral Chamizo MA, Gómez Gracia I, López Montilla D, Roldán Molina R, Collantes Estévez E. Fibrodisplasia osificante progresiva. Descripción de un caso. Rev Osteoporos Metab Miner 2010;3:22. [21] INE. INEbase/Demografía y población/Movimiento Natural de la Población. available from: http://www.ine.es/daco/daco42/mnp/datmnp.htm. [22] Kaplan FS, Xu M, Seemann P, Connor JM, Glaser DL, Carroll L, et al. Classic and atypical fibrodysplasia ossificans progressiva (FOP) phenotypes are caused by mutations in the bone morphogenetic protein (BMP) type I receptor ACVR1. Hum Mutat 2009;30: 379–90. [23] Rogers JG, Geho WB. Fibrodysplasia ossificans progressiva. a survey of forty-two cases. J Bone Joint Surg Am 1979;61-A:909–14. [24] Cohen RB, Hahn GV, Tabas JA, Peeper J, Levitz CL, Sando A, et al. The natural history of heterotopic ossification in patients who have fibrodysplasia ossificans progressiva. A study of forty-four patients. J Bone Joint Surg Am 1993;75:215–9.

755

[25] Kaplan FS, Glaser DL, Shore EM, Deirmengian GK, Gupta R, Delai P, et al. The phenotype of fibrodysplasia ossificans progressiva. Clin Rev Bone Miner Metab 2005;3: 183–8. [26] Connor JM, Evans DAP. Fibrodysplasia ossificans progressiva. The clinical features and natural history of 34 patients. J Bone Joint Surg Br 1982;64:76–83. [27] Bocciardi R, Bordo D, Di Duca M, Di Rocco M, Ravazzolo R. Mutational analysis of the ACVR1 gene in Italian patients affected with fibrodysplasia ossificans progressiva: confirmations and advancements. Eur J Hum Genet 2009;17: 311–8. [28] Scott C, Urban M, Arendse R, Dandara C, Beighton P. Fibrodysplasia ossificans progressiva in South Africa. difficulties in management in a developing country. J Clin Rheumatol 2011;17:37–41. [29] Smith R, Athanasou NA, Vipond SE. Fibrodysplasia (myositis) ossificans progressiva: clinicopathological features and natural history. Q J Med 1996;89:445–56. [30] Schroeder HW, Zasloff M. The hand and foot malformations in fibrodysplasia ossificans progressiva. Johns Hopkins Med J 1980;14:73–8. [31] Kaplan FS, Tabas JA, Zasloff MA. Fibrodysplasia ossificans progressiva: a clue from the fly? Calcif Tissue Int 1990;47:117–25. [32] Deirmengian GK, Hebela NM, O'Connell M, Glaser DL, Shore EM, Kaplan FS. Proximal tibial osteochondromas in patients with fibrodysplasia ossificans progressiva. J Bone Joint Surg Am 2008;90:366–74. [33] Ehara S. Proximal tibial exostosis and fibrodysplasia ossificans progressiva. J Bone Joint Surg Am 2008;90 [letter to Editor]. [34] Kaplan FS, Xu M, Glaser DL, Collins F, Connor M, Kitterman J, et al. Early diagnosis of fibrodysplasia ossificans progressiva. Pediatrics 2008;121:e1295–300. [35] Pignolo RJ, Shore EM, Kaplan FS. Fibrodysplasia ossificans progressiva: clinical and genetic aspects. Orphanet J Rare Dis 2011;6:80. [36] Sun Y, Xia W, Jiang Y, Xing X, Li M, Wang O, et al. A recurrent mutation c.617G>A in the ACVR1 gene causes fibrodysplasia ossificans progressiva in two Chinese patients. Calcif Tissue Int 2009;84:361–5. [37] Nakajima M, Haga N, Takikawa K, Manabe N, Nishimura G, Ikegawa S. The ACVR1 617G>A mutation is also recurrent in three Japanese patients with fibrodysplasia ossificans progressiva. J Hum Genet 2007;52:473–5. [38] Lee DY, Cho T-J, Lee HR, Park MS, Yoo WJ, Chung CY, et al. ACVR1 gene mutation in sporadic Korean patients with fibrodysplasia ossificans progressiva. J Korean Med Sci 2009;24:433–7. [39] Shaw MW, Falls HF, Neel JV. Congenital aniridia. Am J Hum Genet 1960;12:389–415. [40] Lynas MA. Marfan's syndrome in Northern Ireland: an account of thirteen families. Ann Hum Genet 1958;22:289–301. [41] Panchin AY, Mitrofanov SI, Alexeevski AV, Spirin SA, Panchin YV. New words in human mutagenesis. BMC Bioinformatics 2011;12:268–74. [42] Choi SK, Yoon SR, Calabrese P, Arnheim N. Positive selection for New disease mutations in the human germline: evidence from the heritable cancer syndrome multiple endocrine neoplasia type 2B. PLoS Genet 2012;8:e1002420. [43] Goriely A, McVean GAT, Rojmyr M, Ingermarsson B, Wilkie AOM. Evidence for selective advantage of pathogenic FGFR2 mutations in the male germ line. Science 2003;301: 643–6. [44] Goriely A, Wilkie AOM. Missing heritability: paternal age effect mutations and selfish spermatogonia. Nat Rev Genet 2010;11:589. [45] Ratbi I, Bocciardi R, Ragragui A, Ravazzolo R, Sefiani A. Rarely occurring mutation of ACVR1 gene in Moroccan patient with fibrodysplasia ossificans progressiva. Clin Rheumatol 2010;29:119–21. [46] Furuya H, Ikezoe K, Wang L, Ohyagi Y, Motomura K, Fuji N, et al. A unique case of fibrodysplasia ossificans progressiva with an ACVR1 mutation, G356D, other than the common mutation (R206H). Am J Med Genet A 2008;146:459–63. [47] Gregson CL, Hollingworth P, Williams M, Petrie KA, Bullock AN, Brown MA, et al. A novel ACVR1 mutation in the glycine/serine-rich domain found in the most benign case of a fibrodysplasia ossificans progressiva variant reported to date. Bone 2011;48: 654–8. [48] Kaplan FS, Lounev VY, Wang H, Pignolo RJ, Shore EM. Fibrodysplasia ossificans progressiva: a blueprint for metamorphosis. Ann N Y Acad Sci 2011;1237:5–10.