Kaplan, Frederick S. MD; Glaser, David L. MD; Hebela, Nader MD; Shore, Eileen M. PhD
The human skeleton is a complex organ system consisting of more than 200 articulated bones of various shapes and sizes. During embryogenesis, the skeleton develops from undifferentiated mesenchyme according to a genetic plan that controls its precise temporal and spatial formation. Postnatally, the initiation of new skeletal elements is normally restricted to the regeneration of bone at fracture sites.
Heterotopic ossification, the formation of normal bone in an abnormal, soft‐tissue location, results from an alteration in the normal regulation of skeletogenesis.1,2 The formation of bone where it is neither needed nor wanted ranges from clinically nonsignificant coincidental radiographic findings to devastating clinical conditions that dramatically affect quality of life. The causes of heterotopic ossification are numerous and include soft‐tissue trauma, central nervous system injury, and vasculopathies2 (Table 1).
Common Causes of Heterotopic Ossification
Myositis ossificans traumatica may develop after intramuscular hematoma from a sports‐related injury. The molecular pathology is unknown. The ossification process is predominantly endochondral and, in its early stages, may be mistaken for extraosseous osteosarcoma. Sarcomas, however, exhibit the most aggressive histopathologic changes at the periphery of the lesion, whereas immature heterotopic ossification exhibits the most aggressive histopathologic changes at the center of the lesion. Heterotopic ossification also is commonly seen after spinal cord or closed head injury. Attempts to isolate local or systemic inductive factors have not been productive. Intravascular heterotopic ossification may occur in areas of calcified aortic plaques or in cardiac valves.3 Arterial ossification and cardiac valve ossification appear to be highly regulated processes, possibly mediated by bone morphogenetic proteins (BMPs) and pericytelike cells of the vascular wall.3
Heterotopic ossification is a complication of total hip arthroplasty, with the rate of occurrence varying considerably.4 Commonly cited contributing factors include male gender, proliferative osteoarthritis, ankylosing spondylitis, and diffuse idiopathic skeletal hyperostosis.2,4 Again, the molecular pathology is unknown.
A related condition is orthotopic ossification (bone formation limited to ligaments). Ossification of spinal ligaments commonly occurs in patients who have seronegative spondyloarthropathies and may occur in association with diffuse idiopathic skeletal hyperostosis. The pathogenesis of heterotopic ossification is unknown in all of these more common forms as well as in ossification of the posterior longitudinal ligament, an orthotopic spinal ossification common in Japanese men.5
Pathogenesis of Heterotopic Ossification
Aside from the common causes of heterotopic ossification, there are rare and illuminating heritable causes. Regardless of the molecular or genetic etiology, the pathogenesis of heterotopic ossification involves three requisite components: inductive signaling pathways, inducible osteoprogenitor cells, and a heterotopic environment conducive to osteogenesis. Numerous genes are involved in osteogenesis, but few have been implicated in skeletal induction.6 However, analysis of genetic diseases in which skeletal induction is specifically dysregulated has identified some of the genes responsible for and the pathways underlying heterotopic ossification. Two such diseases are fibrodysplasia ossificans progressiva (FOP) and progressive osseous heteroplasia (POH).7,8 These two conditions, in which soft connective tissue is replaced by mature heterotopic bone, differ in the pathologic characteristics of osteogenic induction, histopathologic features of heterotopic osteogenesis, anatomic distribution of heterotopic lesions, and patterns of disease progression.
Although recently researchers have learned much about signaling pathways in the development of the vertebrate limb bud, very little is known about their role in postnatal heterotopic ossification. FOP and POH provide a unique opportunity to study the role of diffusible morphogens, morphogen antagonists, transmembrane receptors, receptorcoupled membrane signaling proteins, downstream transcription factors, and adult mesenchymal stem cells in the earliest events of endochondral and intramembranous bone formation. From such an understanding, more rational and effective prevention and treatment for all forms of heterotopic ossification are likely to arise.
Clinical and Laboratory Findings in Heterotopic Ossification
The symptoms, signs, and laboratory findings are similar in most forms of heterotopic ossification (Table 2). Local symptoms usually include pain, swelling, and decreased mobility of adjacent joints. Early lesions often are mistaken for cellulitis, infection, thrombophlebitis, tumor, or softtissue nonosseous calcification. A detailed medical history typically reveals distinguishing clues that can help to confirm or exclude disorders unrelated to heterotopic ossification.2
Serum calcium and phosphorus levels are normal in all forms of heterotopic ossification and further exclude disorders of mineralization. The serum alkaline phosphatase level, a marker of osteoblast activity, is elevated early in the course of heterotopic ossification but returns to normal as maturation proceeds. Radionuclide bone scans are sensitive but nonspecific and show dramatic increased uptake early in the course of heterotopic ossification before mineralization is apparent on plain radiographs. Biopsy may be helpful in excluding an ossifying soft‐tissue tumor.2 Biopsy, however, often exacerbates heterotopic ossification, especially in patients who have genetic forms of heterotopic ossification such as FOP or POH.7,8
Fibrodysplasia Ossificans Progressiva
FOP is a rare genetic disorder of connective tissue characterized by progressive, disabling ectopic ossification in a characteristic anatomic pattern7 (Fig. 1, Table 3). Congenital malformation of the great toes is the earliest phenotypic feature of FOP and is present in nearly all affected individuals (Fig. 2). Progressive heterotopic ossification often begins after a soft‐tissue injury in the first 5 years of life, but the time of onset is variable. Impending heterotopic ossification is heralded by the rapid appearance of large painful swellings of highly vascular fibroproliferative tissue involving tendons, ligaments, fascia, and skeletal muscle. These preosseous swellings progress along a pathway of endochondral ossification to form mature heterotopic bone. Heterotopic ossification in FOP progresses in specific anatomic and temporal patterns. Typically, the dorsal, axial, cranial, and proximal regions of the body are involved early, and the ventral, appendicular, caudal, and distal regions are involved later (Fig. 1).
Progressive episodes of heterotopic ossification lead to ankylosis of all major joints of the axial and appendicular skeleton, rendering movement impossible. Although the rate of disease progression is variable, most patients are confined to a wheel chair by their early 20s and require lifelong assistance with activities of daily living. Patients with FOP usually succumb later in adulthood to cardiopulmonary complications secondary to severe restrictive pulmonary disease.7 Falls, soft‐tissue injury, surgical trauma, routine intramuscular injections, and various viral illnesses can exacerbate local ossification.7 At present, there is no effective prevention or treatment.
FOP is an autosomal dominant disorder, although most cases are attributable to spontaneous new mutations in previously unaffected families. The genetic defect and pathophysiology of the disorder are not known; however, genes in the BMP pathway have been implicated. Karyotypic abnormalities have not been detected in patients with the disorder, and lesional tissue is not readily available for study. Although a similar condition has been described in domestic cats, no other living animals with the condition are known to exist.7
Histologic examination of early FOP lesions reveals an intense perivascular lymphocytic infiltrate followed by lymphocyte‐associated death of skeletal muscle and robust development of fibroproliferative tissue with extensive neovascularity and mast cell infiltration.9,10 Tissues from FOP lesions at later stages of maturation exhibit characteristic features of endochondral ossification that support ectopic hematopoiesis. Fractures through heterotopic bone appear to heal normally.
The cellular origin of the preosseous fibroproliferative tissue in FOP lesions remains unknown, but muscle satellite cells, endothelial cells, perivascular cells, or mesenchymal stem cells in skeletal muscle or bone marrow are possibilities.2,11‐13 Preliminary data indicate intense immunostaining of fibroproliferative cells of the early lesions with antibodies to BMP4 and numerous smooth muscle proteins. To address the origin of the responding cells, studies are under way using animal models of BMP‐induced heterotopic ossification. In these studies, populations of adult stem cells and progenitor cells are genetically labeled and can be traced after recombinant human BMP4 is injected into skeletal muscle.14
The Role of BMPs in FOP
Because FOP is rare and few multigenerational families exist, standard genetic linkage studies are not feasible. Therefore, the candidate gene approach was used to identify genes that could have an altered expression and function in FOP. The BMPs were primary candidates because this family of proteins can induce mesenchymal stem cells to differentiate to bone through an endochondral pathway in a process that closely parallels the progression of bone formation in an FOP lesion. Examination of the expression levels of numerous BMP genes revealed that only the expression of BMP4 was increased in FOP cells and that these elevated levels were caused by an increased rate of transcription of the BMP4 gene.1,15
With the discovery of altered BMP4 expression in FOP, the BMP4 gene was screened in that population for DNA sequence mutations. Extensive analysis revealed no mutations, a finding supported by the absence of genetic linkage of the BMP4 locus with FOP in four small multigenerational families worldwide.16 Although the BMP4 gene does not harbor the genetic mutation that causes FOP, BMP4 overexpression almost certainly induces formation of bone in this disorder. Currently, differential gene expression and somatic cell mitotic recombination studies are being conducted to identify altered cellular pathways and the mutated gene in FOP that leads to activation of the BMP4 pathway.
The Role of BMP‐BMP Antagonist Interactions
Skeletogenesis, bone induction, and joint formation are specified in a dose‐dependent fashion by tightly regulated morphogenetic gradients, established in part by BMPs and their secreted antagonists.17,18 In embryonic skeletal formation and postnatal bone formation, BMPs restrict their own activity by inducing secreted antagonists in responsive cells. Recent studies have indicated that BMP4 can upregulate expression of multiple secreted BMP antagonists, thereby establishing an autoregulatory negative feedback loop.19
Based on clinical experience with FOP and emerging knowledge of BMP‐BMP antagonist interactions, the hypothesis was formulated that a defect in the autoregulatory feedback pathway between BMP4 and one or more of its extracellular antagonists might contribute to elevated BMP4 activity in FOP.20 Investigation of the basal and BMP4 ‐induced expression of the secreted BMP antagonists (noggin, chordin, and gremlin mRNAs) in control and FOP cell lines has shown that FOP cells exhibit a markedly attenuated upregulation of noggin and gremlin mRNA in response to BMP4 signaling.20 Neither the noggin nor the gremlin genes are mutated in FOP, and those genes also have been excluded by linkage analysis.21
These preliminary observations on the relationship between BMP4 and its secreted antagonists in FOP cells hint at the disease mechanism of FOP: heterotopic ossification in FOP begins in childhood and can be induced by soft‐tissue injury. BMP4, which is expressed by lesional lymphocytes and skeletal muscle, is increased at sites of soft‐tissue injury. Under normal conditions, BMP4 upregulates the expression of its secreted antagonists. A blunted BMP4 antagonist response after soft‐tissue trauma would permit the rapid expansion of a BMP4 morphogenetic gradient conducive to progressive osteogenesis through an endochondral pathway. The growth of highly vascular preosseous fibroproliferative tissue seen locally in response to BMP overexpression would be magnified when the BMP4 antagonist response is blunted and could explain the bone induction seen during an FOP flare‐up. Lymphocytes arriving at a site of soft‐tissue injury exacerbate the process and intensify the inflammatory process at the leading edge of a rapidly expanding lesion. Over time, a large vascular fibroproliferative mass replaces skeletal muscle and matures through an endochondral process into the highly restrictive extra‐articular ribbons, sheets, and plates of bone that ankylose the joints and render movement impossible. Although this hypothesis is plausible, much more work is necessary to find the primary genetic mutation and elucidate the BMP4 regulatory pathway.
Progressive Osseous Heteroplasia
POH is a distinct autosomal dominant disorder of osteogenesis that was discovered as a consequence of investigating FOP.8 POH is characterized by dermal ossification during infancy and by progressive heterotopic ossification of subcutaneous and deep connective tissue during childhood (Fig. 3). The disorder is distinguished from FOP by the absence of congenital skeletal malformations, the absence of predictable regional patterns of heterotopic ossification, the predominance of intramembranous rather than endochondral ossification, and the presence of heterotopic ossification in skin and subcutaneous fat—findings never seen in FOP8 (Table 3).
The first sign of POH is the appearance of cutaneous plaques of intramembranous ossification during infancy. The plaques eventually coalesce and progress to involve the deeper connective tissues, which eventually results in ankylosis of affected joints and focal growth retardation of involved limbs (Fig. 4). As with patients who have FOP, patients with POH have normal intelligence, pass normal developmental milestones, and lack sustained biochemical or endocrine abnormalities. The long‐term prognosis for patients with POH is uncertain because few live beyond early adulthood. Currently, no preventive treatment is available for children with POH8 (Table 3).
Occasional reports of mild heterotopic ossification in Albright's hereditary osteodystrophy (AHO), and a recent study of two AHO patients with atypically extensive heterotopic ossification, suggested the possibility of a common genetic basis for POH and AHO.22,23 AHO is a complex disorder characterized by developmental dysmorphologies commonly associated with multiple hormone resistance. It is caused by heterozygous inactivating mutations in the GNAS1 gene, resulting in decreased expression or function of the alpha subunit of the stimulatory G protein (Gsα) of adenylyl cyclase.
The hypothesis that GNAS1 mutations cause POH was tested by examining the DNA sequences of GNAS1 exons and exon‐intron boundaries in 18 sporadic or familial cases.24 Heterozygous inactivating GNAS1 mutations were identified in 13 of 18 POH probands. The defective allele in POH is inherited exclusively from fathers. This inheritance pattern is consistent with a model of tissue‐specific imprinting, a process by which one of the two parental alleles (in this case, the maternal allele for GNAS1) is inactivated in the relevant cells of the target tissues. A single family provided direct evidence that the same mutation can cause either POH or AHO and that the phenotype correlates with the parental origin of the mutant allele24 (Table 4).
Paternally inherited inactivating GNAS1 mutations cause POH. This finding has immediate implications. It extends the range of phenotypes derived from haploinsufficiency of GNAS1, supports imprinting as a regulatory mechanism for GNAS1 expression, and suggests that Gsα is a critical negative regulator of osteogenic commitment in nonosseous connective tissues, especially in skin, fat, and skeletal muscle.24
Much has been learned about the basic cellular and molecular biology of FOP and POH in the past 5 years, but key questions remain for each condition (Table 4). For FOP, what is the causative gene? For POH, what is the relevant signaling pathway? For both conditions, what are the responsible osteoprogenitor cells in the responding tissues? Answers to these and related questions will permit a better understanding of the pathophysiology of heterotopic ossification in humans. That understanding likely will lead to a more rational approach to treatment of the genetic disorders as well as of the more common acquired forms of heterotopic ossification.
Prevention and Treatment of Heterotopic Ossification
The prevention and treatment of heterotopic ossification are based on three principles: disrupting the relevant inductive signaling pathways, altering the relevant osteoprogenitor cells in the target tissue or tissues, or modifying the environment conducive to heterotopic osteogenesis. For most forms of heterotopic ossification (both acquired and developmental), the identity of these targets is unknown (Table 4). Nevertheless, important observations and controlled clinical testing have led to promising prevention and treatment modalities. Some of these modalities are currently available; others are in various stages of preclinical and clinical testing.
Disrupting Inductive Signaling Pathways
Inflammatory prostaglandins are potent co‐stimulatory molecules with BMPs in the induction of heterotopic bone.25 Lowering prostaglandin levels in the tissues of experimental animals raises the threshold for heterotopic ossification, thus making it more difficult for heterotopic bone to form.26 Following intramuscular injections of BMP‐containing demineralized bone matrix, animals pretreated with prostaglandin inhibitors, such as indomethacin, form less heterotopic bone than control animals do.
Several recent randomized trials of nonsteroidal anti‐inflammatory drugs (NSAIDs) to prevent heterotopic bone formation after major hip surgery substantiate that NSAIDs are useful prophylactic agents in these patients. However, the regimen, agent, dosage, and duration of prophylaxis vary greatly from study to study.27
Two recent studies delineate the critical role of cyclooxygenase‐2 (COX‐2) in the inflammatory bone regeneration of fracture healing and thus raise the possibility that COX‐2 inhibitors might be useful clinically for heterotopic ossification.28,29 Nevertheless, it is clear that BMPs can act downstream of COX‐2 and might bypass the inhibition of fracture healing that the COX‐2 inhibitors impose.29
Effective prevention and treatment of heterotopic ossification after spinal cord injury or closed head trauma has been problematic, but combinations of anti‐inflammatory drugs (to disrupt the inductive signal), preoperative radiation (to disrupt the responding cells), and postoperative etidronate (to disrupt the conducive environment) may have some beneficial effects. Generally, surgical excision of heterotopic bone has been limited to patients with advanced symptoms of ankylosis and has been delayed until the heterotopic bone is mature, as determined by radiographic and radionuclide studies.2 However, more recent studies have described successful early excision of heterotopic bone followed by radiation therapy, especially about the elbow.30
Gene therapy with BMP antagonists seems to be a promising potential treatment based on present knowledge of the condition. Because the overexpression of BMP4 and the underexpression of multiple BMP antagonists have important implications in the pathogenesis of FOP, researchers have attempted to utilize knowledge gained from these relationships. Glaser et al31 developed a mouse model of induced heterotopic osteogenesis using injectable basement membrane matrix (an osteoconductive material) impregnated with BMP4 (an osteoinductive morphogen). Half of the animals were pretreated with a dose of adenovirus particles containing a modified human noggin gene (noggin is a secreted protein that binds multiple BMPs and prevents them from binding to their transmembrane receptors); half of the animals were pretreated with adenovirus vector alone. Implants were recovered at 7 and 14 days after injection, and standard histologic techniques were used to evaluate the stages of bone formation.
In control animals, the implants containing BMP4 induced an aggressive fibroproliferative lesion with chondrogenesis at 7 days and heterotopic osteogenesis at 14 days. However, in animals treated with noggin‐containing adenovirus, the implants containing BMP4 demonstrated only a thin pseudocapsule at 14 days, indistinguishable from carrier implants with no BMP4. Thus, the delivery of noggin protein through gene therapy successfully prevents BMP4‐induced heterotopic ossification in a mouse model. The study provides proof of the concept that a secreted morphogen antagonist can be produced in vivo and act systemically to prevent BMP4‐mediated heterotopic ossification. Numerous obstacles need to be overcome before noggin gene therapy could be used to treat patients with FOP, including the development of safe and effective viral vectors and the use of inducible promoters to regulate gene activity.
Altering the Population of Progenitor Cells in Target Tissue
Several studies suggest that preoperative radiation effectively prevents heterotopic ossification after total joint arthroplasty, eliminating the discomfort and morbidity of conventional postoperative treatments.32,33 Furthermore, the efficacy of preoperative radiation suggests that the mesenchymal stem cells or osteogenic precursor cells that are active in this process are derived, at least in part, from the local tissues within the surgical field.
Recent advances in basic and clinical research suggest that stem cells may lie at the heart of a cure for FOP.11‐13 Hematopoietic cells have been found in biopsies of lesions, and recently stem cells have been reported to give rise to multiple mesenchymal tissues, including muscle and bone.9,10 Given these new insights, it is rational to ask whether patients with FOP should be treated with stem cell transplantation from bone marrow, peripheral blood, or umbilical cord blood to replace their hematopoietic stem cell pool. To answer this question, it is necessary to consider how stem cell transplantation might cure FOP, how it might fail, and what clinical risks patients might have to accept.34
Modifying the Environment Conducive to Heterotopic Ossification
Modifying the environment conducive to heterotopic ossification may help prevent bone formation in abnormal places. Angiogenesis is an absolute requirement for the formation and development of the skeleton, for the successful healing of fractures, and for the formation of heterotopic bone.35 Angiogenesis, a prominent histopathologic feature of preosseous FOP lesions, becomes a potential target for therapy in this disease. The goal of antiangiogenic therapy in heterotopic ossification is to inhibit new blood vessel formation to slow or inhibit the subsequent production of heterotopic bone.
Sodium etidronate has been used empirically because it inhibits bone matrix mineralization and may impair ossification at high dosages. However, when used at high doses for sustained periods, sodium etidronate causes osteomalacia and can impair ossification of normal bone. Its clinical utility therefore is limited.2
The underlying cause of heterotopic ossification may be the most important consideration in weighing surgical management. Although surgical resection may be beneficial in some cases of posttraumatic or neurologically associated heterotopic ossification, it is ineffective or detrimental in others, especially in patients with the genetic disorders of heterotopic ossification. In FOP, for example, surgically removing mature heterotopic bone is contraindicated because ossification inevitably recurs and the surgery may cause the disease to progress extensively.7 In POH, surgical removal of mature heterotopic bone is not contraindicated. However, it is extremely difficult and may be technically impossible because heterotopic bone is dispersed through the muscle fascicles.8
Preventing soft‐tissue injury and muscle damage, as well as preventing falls, remains a hallmark of FOP management. Intramuscular injections must be avoided assiduously and extraordinary care must be practiced in all dental work because minimal soft‐tissue trauma during routine dental care may induce permanent ankylosis of the jaw. Falls suffered by FOP patients can lead to severe injuries and flare‐ups.7 These admonishments are a sober reminder that, even in an era in which attention is focused on molecular mechanisms of disease, simple measures can prevent catastrophic harm.
Research in the various forms of heterotopic ossification must continue to identify inducible pathways, receptive target cells, and environments conducive to producing heterotopic bone. The two rare genetic conditions FOP and POH provide important starting points for understanding some of the major molecular and genetic pathways involved in heterotopic ossification in humans. It is likely that some of the familiar forms of heterotopic ossification share many of these pathways.
A more complete understanding of heterotopic ossification likely will have therapeutic implications for patients undergoing spinal fusion, peripheral joint fusion, fracture management, and surgery to reconstruct limbs and treat congenital malformations. The basic research that leads to the treatment and prevention of heterotopic ossification in one group of patients could lead to therapeutic advances in tissue engineering that could help others who do not form adequate amounts of bone. Finally, insight gained from the study of rare disorders of heterotopic ossification will enhance our understanding of the normal pathways of bone formation.
Acknowledgments: This work was supported in part by research grants from the International FOP Association, the Progressive Osseous Heteroplasia Association, the Ian Cali Endowment for FOP Research, the Isaac and Rose Nassau Professorship of Orthopaedic Molecular Medicine, the Center for Research in FOP and Related Disorders, and the National Institute of Arthritis and Musculoskeletal and Skin Diseases (2‐RO1‐AR‐41916‐04‐ORTH; 1‐RO1‐AR‐46831‐01‐ORTH).
1. Shafritz AB, Shore EM, Gannon FH, et al: Overexpression of an osteogenic morphogen in fibrodysplasia ossificans progressiva. N Engl J Med
2. Kaplan FS, Hahn GV, Zasloff MA: Heterotopic ossification: Two rare forms and what they can teach us. J Am Acad Orthop Surg
3. Mohler ER III, Gannon F, Reynolds C, Zimmerman R, Keane MG, Kaplan FS: Bone formation and inflammation in cardiac valves. Circulation
4. Ahrengart L: Periarticular heterotopic ossification after total hip arthroplasty: Risk factors and consequences. Clin Orthop
5. Tanaka T, Ikari K, Furushima K, et al: Genomewide linkage and linkage disequilibrium analyses identify COL6A1, on chromosome 21, as the locus for ossification of the posterior longitudinal ligament of the spine. Am J Hum Genet
6. Einhorn TA, Lee CA: Bone regeneration: New findings and potential clinical applications. J Am Acad Orthop Surg
7. Kaplan FS, Shore EM, Connor JM: Fibrodysplasia ossificans progressiva, in Royce PM, Steinmann B (eds): Connective Tissue and its Heritable Disorders
: Molecular, Genetic, and Medical Aspects
, ed 2. New York, NY: Wiley-Liss, 2002, pp 827-840.
8. Kaplan FS, Shore EM: Progressive osseous heteroplasia. J Bone Miner Res
9. Gannon FH, Valentine BA, Shore EM, Zasloff MA, Kaplan FS: Acute lymphocytic infiltration in an extremely early lesion of fibrodysplasia ossificans progressiva. Clin Orthop
10. Gannon FH, Glaser D, Caron R, Thompson LD, Shore EM, Kaplan FS: Mast cell involvement in fibrodysplasia ossificans progressiva. Hum Pathol
11. Bianco P, Robey PG: Stem cells in tissue engineering. Nature
12. Bosch P, Musgrave DS, Lee JY, et al: Osteoprogenitor cells within skeletal muscle. J Orthop Res
13. Nilsson SK, Dooner MS, Weier HU, et al: Cells capable of bone production engraft from whole bone marrow transplants in nonablated mice. J Exp Med
14. Glaser DL, Ramachandran R, Jiao X, et al: Abstract: The heritage of cells within postnatal tendon and bone formation. J Bone Miner Res
15. Roush W: Protein builds second skeleton. Science
16. Feldman G, Li M, Martin S, et al: Fibrodysplasia ossificans progressiva, a heritable disorder of severe heterotopic ossification, maps to human chromosome 4q 27-31. Am J Hum Genet
17. Brunet LJ, McMahon JA, McMahon AP, Harland RM: Noggin, cartilage morphogenesis, and joint formation in the mammalian skeleton. Science
18. Perrimon N, McMahon AP: Negative feedback mechanisms and their roles during pattern formation. Cell
19. Gazzerro E, Gangji V, Canalis E: Bone morphogenetic proteins induce the expression of noggin, which limits their activity in cultured rat osteoblasts. J Clin Invest
20. Ahn J: Serrano de la Pena L, Shore EM, Kaplan FS: Paresis of a bone morphogenetic protein-antagonist response in a genetic disorder of heterotopic skeletogenesis. J Bone Joint Surg Am
21. Xu M-Q, Feldman G, Le Merrer M, et al: Linkage exclusion and mutational analysis of the noggin gene in patients with fibrodysplasia ossificans progressiva (FOP). Clin Genet
22. Eddy MC, Jan De Beur SM, Yandow SM, et al: Deficiency of the alphasubunit of the stimulatory G protein and severe extraskeletal ossification. J Bone Miner Res
23. Yeh GL, Mathur S, Wivel A, et al: GNAS1 mutation and Cbfa
24. Shore EM, Ahn J, Jan de Beur S, et al: Paternally inherited inactivating mutations of the GNAS1 gene in progressive osseous heteroplasia. N Engl J Med
25. Ono I, Inoue M, Kuboki Y: Promotion of the osteogenetic activity of recombinant human bone morphogenetic protein by prostaglandin E1. Bone
26. DiCesare PE, Nimni ME, Peng L, Yazdi M, Cheung DT: Effects of indomethacin on demineralized bone-induced heterotopic ossification in the rat. J Orthop Res
27. Neal BC, Rodgers A, Clark T, et al: A systematic survey of 13 randomized trials of non-steroidal anti-inflammatory drugs for the prevention of heterotopic bone formation after major hip surgery. Acta Orthop Scand
28. Simon AM, Manigrasso MB, O'Connor JP: Cyclo-oxygenase 2 function is essential for bone fracture healing. J Bone Miner Res
29. Zhang X, Schwarz EM, Young DA, Puzas JE, Rosier RN, O'Keefe RJ: Cyclooxygenase-2 regulates mesenchymal cell differentiation into the osteoblast lineage and is critically involved in bone repair. J Clin Invest
30. McAuliffe JA, Wolfson AH: Early excision of heterotopic ossification about the elbow followed by radiation therapy. J Bone Joint Surg Am
31. Glaser DL, Economides AN, Wang L, et al: In vivo somatic cell gene transfer of an engineered Noggin mutein prevents BMP4-induced heterotopic ossification. J Bone Joint Surg Am
32. Burd TA, Lowry KJ, Anglen JO: Indomethacin compared with localized irradiation for the prevention of heterotopic ossification following surgical treatment of acetabular fractures. J Bone Joint Surg Am
33. D'Lima DD, Venn-Watson EJ, Tripuraneni P, Colwell CW: Indomethacin versus radiation therapy for heterotopic ossification after hip arthroplasty. Orthopedics
34. Emerson SG, Kaplan FS: Answer: Marrow stem cell transplantation in fibrodysplasia ossificans progressiva. Clin Orthop
35. Yin M, Gentili C, Koyama E, Zasloff M, Pacifici M: Antiangiogenic treatment delays chondrocyte maturation and bone formation during limb skeletogenesis. J Bone Miner Res