Abstract
Melorheostosis is a rare sclerosing bone disease characterized by excessive cortical bone deposition that is frequently on the differential diagnosis for bone biopsies. Although the radiologic pattern of “dripping candle wax” is well known, the pathologic findings have been poorly defined. Here, we comprehensively describe the histology of melorheostosis in 15 patients who underwent bone biopsies. Common histologic findings included: dense cortical bone (73.3%), woven bone (60%), and hypervascular features and increased porosity (66.7%). One third of the patients (5/15) also had prominent cement lines. Multiple patients had >1 histologic pattern (ie, dense cortical bone and hypervascularity). Overall, this study suggests that melorheostosis exists with several histologically distinct patterns. When confronted with a case of suspected melorheostosis, the clinical pathologist should use the histologic features common to melorheostotic lesions presented here in conjunction with the patient’s clinical presentation and radiographic findings to arrive at a diagnosis. An illustrative case is presented.
Keywords: sclerosing bone dysplasia, hyperostosis, bone biopsies, histology, osteoidosis
Melorheostosis (OMIM%155950), a type of sclerosing bone disease, is an extremely rare skeletal abnormality that affects both cortical bone and surrounding soft tissue.1 Melorheostosis was first described in the literature nearly a century ago,2 yet its pathologic findings have been poorly described. Similar to other sclerosing dysplasias, patients present with hyperostotic features that most often manifest as pain at the affected site. Melorheostosis classically manifests in adolescence or young adulthood with a prevalence of 0.9 per million.3
The diagnosis of melorheostosis remains a challenge, as it can present with clinical and radiologic features similar to tumors and other sclerosing bone diseases, such as intramedullary osteosclerosis and osteopoikilosis.4 Although its characteristic radiographic “dripping candle wax” appearance aids in the diagnosis, this is not a universal finding in patients with melorheostosis. Therefore, the lack of a comprehensive pathologic description and a confirmatory test can make the diagnosis subjective. A recent area of interest in the study of melorheostosis has centered on the genetic mutations seen in these patients. In 2018, our group identified somatic activating mutations in the regulatory domain of the MAP2K1 kinase,5,6 which underlies the pathologic changes for many patients with the “dripping candle wax”finding.
As the understanding of melorheostosis continues to evolve, a holistic picture of the disease is a necessary step in this process. More specifically, a better description of the pathologic features will aid in the definitive diagnosis of this disorder. To date, abnormal tissue found in melorheostotic lesions has not been well defined. Thus, the aim of this study is to comprehensively describe the pathologic findings seen in a cohort of patients diagnosed with melorheostosis and compare these findings to healthy bone. A secondary objective is to describe the clinical and radiologic features in these patients, which will aid the clinical pathologist presented with a case of possible melorheostosis.
MATERIALS AND METHODS
From 2015 to 2018, a cohort of 30 patients with melorheostosis enrolled in an Institutional Review Board-approved natural history study in the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) (Clinicaltrials.gov NCT02504879). All participants provided written informed consent.
Eligible participants met the following criteria: (1) 18 years or above; (2) suspected or diagnosed with melorheostosis; and (3) radiographic evidence consistent with melorheostosis. A diagnosis of melorheostosis was confirmed by correlating radiographic findings with anatomic areas of increased uptake on fluorine-18 sodium fluoride positron emission tomography/computed tomography (CT) studies. Increased fluorine-18 sodium fluoride activity is indicative of increased blood supply to the bone, as well as increased turnover.7
Fifteen patients consented for paired biopsies of affected and unaffected bone. Each patient had an open surgical biopsy of the affected melorheostotic lesion, as well as a biopsy of contralateral, unaffected bone. For the biopsies from affected bone, specimen size was variable depending on the degree of pathologic involvement. Four-micron-thick hematoxylin and eosin (H&E)–stained sections underwent histologic evaluation. These tissue sections were formalin-fixed, processed and paraffin-embedded. Other sections were fixed in 70% ethanol, deplasticized with 2-methoxyethyl acetate and stained with modified Goldner Trichrome8 and viewed under transmission and polarized light microscopy. DNA was extracted from a portion of affected and unaffected bone for genetic testing. Digital images were captured with a camera (AxioCam HRc; Zeiss, Oberkochen, Germany) attached to a light microscope (Axiophot; Zeiss).
RESULTS
Clinical information for the entire cohort is reported in Table 1. In our 15 biopsied patients, there were 10 females and 5 males. The mean age at the time of biopsy was 45.3 ± 12.3 years, and patients had an average disease duration of 29.0 ± 14.8 years. To paint an overall picture of melorheostosis, the clinical, radiologic, and pathologic findings of 1 patient from the cohort, Melo-09, will be highlighted in addition to reporting findings for the entire cohort.
TABLE 1.
Clinical Features
| Case | Age (y) | Age of Diagnosis (y) | Disease Duration (y) | Sex | Radiology | Histologic Patterns | Extraosseous Mineralization | Polyostotic Disease | MAP2K1 Status |
|---|---|---|---|---|---|---|---|---|---|
| Melo-01 | 26 | 12 | 14 | Female | Endosteal | Dense cortical bone, woven bone | No | Yes | − |
| Melo-02 | 42 | 10 | 32 | Male | Classic | Hypervascularity, invasion of the medullary cavity, prominent cement lines | Yes | Yes | + |
| Melo-04 | 41 | 10 | 31 | Female | Classic | Dense cortical bone, hypervascularity, periosteal fibrous thickening, prominent cement lines, woven bone | Yes | Yes | + |
| Melo-08 | 52 | 10 | 42 | Female | Endosteal | Dense cortical bone, increased porosity, woven bone | No | Yes | − |
| Melo-09 | 59 | 12 | 47 | Female | Classic | Increased porosity, woven bone | No | Yes | + |
| Melo-10 | 49 | 8 | 41 | Female | Classic | Dense cortical bone, hypervascularity, woven bone | Yes | Yes | + |
| Melo-11 | 65 | 60 | 5 | Male | Endosteal | Hypervascularity, woven bone | No | No | − |
| Melo-12 | 30 | 5 | 25 | Female | Endosteal | Hypervascularity, prominent cement lines | No | Yes | − |
| Melo-13 | 57 | 18 | 39 | Female | Endosteal | Dense cortical bone | Yes | Yes | − |
| Melo-16 | 27 | 24 | 3 | Female | Classic | Dense cortical bone | No | Yes | + |
| Melo-17 | 35 | 5 | 30 | Male | Endosteal | Dense cortical bone, prominent cement lines, woven bone | No | Yes | − |
| Melo-18 | 45 | 10 | 35 | Female | Classic | Dense cortical bone, hypervascularity, woven bone | Yes | Yes | + |
| Melo-19 | 41 | 33 | 8 | Male | Classic | Dense cortical bone, invasion of the medullary cavity | Yes | Yes | + |
| Melo-22 | 50 | 13 | 37 | Female | Mixed* | Dense cortical bone, hypervascularity | Yes | Yes | + |
| Melo-24 | 60 | 14 | 46 | Male | Classic | Dense cortical bone, increased porosity, invasion of the medullary cavity, prominent cement lines, woven bone | No | Yes | |
Mixed = classic+myositis ossificans-like.
indicates that MAP2K1 mutation was present in this patient
, mutation not present; MAP2K1, mitogen-activated protein kinase kinase 1.
Clinical Features
Patient Melo-09 was a 59-year-old female at the time of her tissue biopsy procedure. She was 12 years old when she first experienced pain symptoms. Her primary lesion was in the right upper extremity involving the hand; she did not have any extraosseous lesions (Fig. 1). On physical examination, she had decreased sensation at the lesion. She did not have any skin findings overlying the primary bone lesion.
FIGURE 1.
Clinical manifestations of melorheostosis. A, A 59-year-old female (Melo-09) with irregular bone growth of her right hand remarkable for swelling of the right thumb. B, Radiograph showing classic candle wax appearance of the thumb and index finger. C, Magnetic resonance imaging of the right hand notable for the absence of an increased signal on STIR. D, Lateral radiograph of the elbow revealing melorheostosis of the anterior aspect of the humerus.
Radiologic Findings
The radiographic pattern present in Melo-09 was the classic “dripping candle wax” appearance (Figs. 1B, D). In the group of biopsied patients, 53.3% also had the classic radiographic findings. The fluorine-18 sodium fluoride positron emission tomography/CT scan of Melo-09 demonstrated increased activity.
The affected lesion in Melo-09 was positive for the MAP2K1 mutation (Table 1). Of the entire population, 8/15 (53%) were MAP2K1 positive. No other genetic mutations (ie, LEMD3) were present.
Pathologic Findings
The histologically distinct patterns from cortical bone sections from different patients are presented in Figure 2. Figure 2A shows a section with nearly normal bone in Melo-18. In contrast, within the entire cohort, histology findings on H&E staining varied from patient to patient and within lesions (Fig. 2). Histologic examination of melorheostotic bone in Melo-09 reveals woven bone with 2 to 3 times more Haversian canals per high power field than normal bone, which represents increased porosity (Fig. 2B). Dense cortical bone transiting to the cartilaginous bone and very irregular structure is best appreciated in Figures 2C and D.
FIGURE 2.
Histologic spectrum of melorheostosis viewed in hematoxylin and eosin-stained sections. Images progress from near-normal bone histology to an extremely disorganized bone structure seen in melorheostosis. A, Appositional bone with a Haversian canal traversing perpendicular to the plane of the appositional bone (Melo-18). Thickened cortical bone with prominent cement lines is visualized (black arrows). B, More irregular woven bone (Melo-09). C, Dense cortical bone transitioning to woven bone transitioning to fibrocartilage (Melo-04). D, Highly irregular bone (Melo-18). Bone in this image is extremely dense, as the section could not be cut any thinner. Melo-18 has a bone that appears differently in different areas of tissue.
Histologic patterns for the entire cohort were tabulated from pathology reports. The most common histologic pattern in affected samples was dense cortical bone, which was seen in nearly three fourths of the population (11/15 patients). Overall, 60% (9/15) of affected samples had woven bone features present, which is best appreciated under polarized light (Fig. 3B). About half of the patients (7/15) demonstrated hypervascular features and another 3 patients had increased porosity, both indicated by an increased number of Haversian systems. One third of the patients (5/15) had prominent cement lines. Irregular bone growth into the medullary cavity was appreciated in 20% (3/15) patients, whereas only 1 patient had periosteal fibrous thickening (Melo-04). Multiple patients had >1 histologic pattern (ie, dense cortical bone and hypervascularity).
FIGURE 3.
Histologic images of affected bone stained with modified Goldner Trichrome viewed under transmission and polarized light (Melo-04). A, Transmission light image showing regions of organized parallel lamellar bone (black asterisks) and woven bone (black arrows). Large vascular channels with varying amounts of osteoid deposition (red) are also appreciated. B, Identical section under polarized light, which better highlights the zones of parallel lamellar bone (white asterisks) and zones of woven bone (white arrows).
Varying amounts of newly deposited unmineralized osteoid were seen in affected lesions stained with modified Goldner Trichrome and viewed either under transmission light (Figs. 3A, 4A–D) or under polarized light (Fig. 3B) for better visualization of lamellar orientation. Although Melo-02 had a large amount of newly deposited osteoid (Fig. 4C), Melo-10 had little osteoid in the present image (Figs. 4A, B). However, considering the total biopsy sample, osteoid indices were highly increased as previously shown by histomorphometric evaluations.5 Even within the same patient (Melo-10), histologic sections had a spectrum of patterns with respect to osteoid deposition (Figs 4A, B).
FIGURE 4.
Histologic sections stained with modified Goldner Trichrome viewed under transmission light. As previously reported, melorheostotic lesions consist of new primary bone deposition by the periosteum, which becomes subsequently remodeled into osteonal-like bone, with varying amounts of newly deposited osteoid.5,6,9 All histologic sections are from the affected bone with red showing unmineralized osteoid and green depicting mineralized bone. A, Bone tissue porosity is highly increased with the variable osteoid present (Melo-10). B, Scant osteoid present (Melo-10). C, Prominent osteoid deposition surrounding large vascular channels, which interrupts the organized parallel lamellar bone (Melo-02). D, Large vascular channel with prominent surrounding osteoid and active osteoblasts (black arrows in C and D) (Melo-04).
DISCUSSION
Melorheostosis is exceptionally rare bone dysostosis that can be challenging to diagnose. One factor that contributes to the diagnostic difficulty is the very low prevalence of melorheostosis. The aim of this paper is to aid the clinical pathologist to recognize and diagnose melorheostosis.
Most cases of melorheostosis present in adolescents or young adults with a similar clinical picture.1 This held true in our cohort, in which the average age was 16.3 ± 14.1 years at the time of first symptoms. The most common first symptoms experienced by these patients were a new mass accompanied by pain. Initial symptoms primarily revolve around bone pain and joint stiffness, swelling, and deformity of the limb.10,11 Melorheostosis mostly affects the long bones of the upper and lower extremities in a unilateral fashion; less commonly, the axial skeleton is involved.1,11,12
Melorheostosis, characterized by dense hyperostotic lesions, is typically diagnosed in patients with islands of radiodensity in the long bones of the appendicular skeleton.6 The classic radiologic appearance of the lesion is “dripping candle wax.5” However, other radiographic features have also been described in patients with melorheostosis, including myositis ossificans-like, osteoma-like (endosteal), osteopathia striata-like, and a mixed pattern.10 Over 50% of this cohort of patients (8/15, 53.3%) had the classic radiographic appearance, whereas 6/15 (40.0%) had an endosteal radiographic pattern.
The majority of patients presenting with melorheostotic features can be diagnosed based on their radiologic findings (x-rays, CT, magnetic resonance imaging, and bone scans).13,14 However, as noted above, there is no one specific radiographic finding for melorheostosis, and the differential diagnosis may be broad even after imaging. Therefore, a biopsy may be warranted in the cases of perplexing or suspicious lesions to achieve a definitive diagnosis.
Although there are no pathognomonic histologic features of melorheostosis, a bone biopsy of an undefined lesion is a necessary step to rule out potential malignancy. Melorheostosis is a benign disease,15 and malignant features like spindle-shaped cells16 are not present. In general, melorheostotic bone histology shows a periosteal reaction with parallel lamellar bone apposition and subsequent diffuse remodeling,5,6,9 which is not present in unaffected tissue. More specifically, Hoshi et al17 showed an increase in disorganized osteoid formation with increased angiogenesis. Although having a broad overlying histologic picture can be helpful, it is important to understand the histologic variation seen in a rare disease like melorheostosis. In this cohort of patients, the histologic findings ranged from simple dense cortical bone to highly woven bone. The most frequent histologic findings in tissue samples from this patient cohort were dense cortical bone, woven bone, hypervascular features including increased porosity, and prominent cement lines. When investigating a suspected case of melorheostosis on H&E staining, the pathologist should look for these findings. On the basis of this cohort, the presence or absence of osteoblastic or osteoclastic activity would not aid in the diagnosis. Genetic testing is not yet commercially available and is made more complex by somatic mosaicism in the lesions.
The histologic variability seen in affected tissue samples may depend on certain factors related to the biopsy. First, the site of biopsy may dictate the histology seen on H&E. For example, the histology of the tissue biopsy in patient Melo-04 showed 3 distinct patterns: dense cortical bone, woven bone, and fibrocartilage. If the biopsy was taken in a slightly different location, the tissue seen on H&E may have only revealed dense cortical bone or woven bone. Second, the depth of the biopsy taken may also change the histology seen by the pathologist. Previous studies have shown that, in melorheostotic lesions, there are 2 drastically different regions: an outer capsule of bone with minimal porosity and an inner region of more porous bone.5,6,9 Depending on the depth of biopsy, tissue containing a little or a great deal of porosity may be captured.
Overall, our findings suggest that melorheostosis is histologically variable, with a number of common features. Thus, identifying certain histologic patterns commonly found in melorheostotic lesions like increased cortical density, woven bone, increased osteoid deposition and increased vascularity is an important step in the diagnosis. Because of its limited prevalence, most pathologists do not frequently see melorheostosis; however, using clinical, radiologic, and histologic data presented here will aid the clinical pathologist in distinguishing melorheostosis from other sclerosing bone diseases.
ACKNOWLEDGMENTS
The authors thank the Laboratory of Pathology at the National Cancer Institute (NCI) and the Bone Laboratory of the Ludwig Boltzmann Institute of Osteology in Vienna, Austria (Daniela Gabriel, Petra Keplinger, Sonja Lueger, and Phaedra Messmer) for careful sample preparations.
Conflicts of Interest and Source of Funding: Supported by the Division of Intramural Research of the National Institute of Arthritis and Musculoskeletal and Skin Diseases, the Eunice Kennedy Shriver National Institute of Child Health and Human Development, the Austrian Workers’ Compensation Board/Austrian Social Insurance for Occupational Risks (AUVA), and the Vienna Regional Health Insurance Fund (WGKK). C.N.F. acknowledges funding support through the National Institutes of Health (NIH) Medical Research Scholars Program, a public-private partnership supported jointly by the NIH and generous contributions from the Doris Duke Charitable Foundation, Genentech, American Association for Dental Research, the Colgate-Palmolive Company, Elsevier, and other private donors.
Footnotes
The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article.
REFERENCES
- 1.Kotwal A, Clarke BL. Melorheostosis: a rare sclerosing bone dysplasia. Curr Osteoporos Rep. 2017;15:335–342. [DOI] [PubMed] [Google Scholar]
- 2.Léri AJJ. An undescribed condition of bones. Hyperostosis in the entire length of a limb or malarostosis [in French]. Bull Mem Soc Med Hop (Paris). 1922;46:1141–1145. [Google Scholar]
- 3.Wynne-Davies R, Gormley J. The prevalence of skeletal dysplasias. An estimate of their minimum frequency and the number of patients requiring orthopaedic care. J Bone Joint Surg Br. 1985;67:133–137. [DOI] [PubMed] [Google Scholar]
- 4.Ihde LL, Forrester DM, Gottsegen CJ, et al. Sclerosing bone dysplasias: review and differentiation from other causes of osteosclerosis. Radiographics. 2011;31:1865–1882. [DOI] [PubMed] [Google Scholar]
- 5.Kang H, Jha S, Deng Z, et al. Somatic activating mutations in MAP2K1 cause melorheostosis. Nat Commun. 2018;9:1390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jha S, Fratzl-Zelman N, Roschger P, et al. Distinct clinical and pathological features of melorheostosis associated with somatic MAP2K1 mutations. J Bone Miner Res. 2019;34:145–156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nawata S, Kaneta T, Ogawa M, et al. Differences in sodium fluoride-18 uptake in the normal skeleton depending on the location and characteristics of the bone. Nuklearmedizin. 2017;56:91–96. [DOI] [PubMed] [Google Scholar]
- 8.Glorieux FH, Travers R, Taylor A, et al. Normative data for iliac bone histomorphometry in growing children. Bone. 2000;26:103–109. [DOI] [PubMed] [Google Scholar]
- 9.Fratzl-Zelman N, Roschger P, Kang H, et al. Melorheostotic bone lesions caused by somatic mutations in MAP2K1 have deteriorated microarchitecture and periosteal reaction. J Bone Miner Res. 2019;34: 883–895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Freyschmidt J Melorheostosis: a review of 23 cases. Eur Radiol. 2001;11:474–479. [DOI] [PubMed] [Google Scholar]
- 11.Smith GC, Pingree MJ, Freeman LA, et al. Melorheostosis: a retrospective clinical analysis of 24 patients at the Mayo Clinic. PM R. 2017;9:283–288. [DOI] [PubMed] [Google Scholar]
- 12.Suresh S, Muthukumar T, Saifuddin A. Classical and unusual imaging appearances of melorheostosis. Clin Radiol. 2010;65:593–600. [DOI] [PubMed] [Google Scholar]
- 13.Artner JCB, Wernerus D, Reichel H, et al. Melorheostosis: current concepts in diagnosis and treatment—a review of literature (313 cases). J Musculoskelet Res. 2012;15:1230002. [Google Scholar]
- 14.Judkiewicz AM, Murphey MD, Resnik CS, et al. Advanced imaging of melorheostosis with emphasis on MRI. Skeletal Radiol. 2001;30: 447–453. [DOI] [PubMed] [Google Scholar]
- 15.Kumar R, Sankhala SS, Bijarnia I. Melorheostosis—case report of rare disease. J Orthop Case Rep. 2014;4:25–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.de Andrea CE, Petrilli AS, Jesus-Garcia R, et al. Large and round nuclei in osteosarcoma: good clinical outcome. Int J Clin Exp Pathol. 2011;4:169–174. [PMC free article] [PubMed] [Google Scholar]
- 17.Hoshi K, Amizuka N, Kurokawa T, et al. Histopathological characterization of melorheostosis. Orthopedics. 2001;24:273–277. [DOI] [PubMed] [Google Scholar]




