We reviewed the literature to determine the important advances related to primary bone tumors during the past few years. In the present update, we will discuss malignant tumors (including Ewing sarcoma, osteosarcoma, chondrosarcoma, and chordoma) and benign tumors (including benign cartilaginous tumors, giant cell tumor of bone, and aneurysmal bone cysts).
With regard to malignant tumors, much effort has been focused on finding new potential targets for systemic therapy. Additional efforts have been made in an attempt to identify disease markers that will correlate with outcome. For benign tumors, the focus has been on less aggressive treatment over time. For example, twenty-five years ago, giant cell tumor of bone was often treated with en bloc excision, whereas now, local curettage with or without an adjuvant is more routine. Presently, drugs that act on osteoclasts, such as denosumab (an antibody to RANK ligand), may result in management of giant cell tumors without surgery particularly with very large or recurrent tumors. We will review some of the newer aspects of detection and treatment of some of the more common primary malignant and benign bone tumors.
The primary goal of treating malignant bone sarcomas, at least those of higher grade, is a negative surgical margin without entry into the tumor or its surrounding reactive zone. Low-grade malignant bone tumors may be less aggressively treated, and the resulting bone defect requires less aggressive treatment.
High-grade tumors usually require aggressive surgical resection, which results in a large osseous defect. The reconstruction of resulting defects has been approached with a variety of methods. Unlike routine joint reconstruction, the osseous defects are quite large and the amount of bone loss quite substantial. What may be even more important is the resulting soft-tissue defect being a major challenge that may even affect the ability to perform limb-salvage surgery.
The reconstructive procedures for large osseous defects vary according to the preference of the surgeon, the surgical site, the disease process, and sometimes the availability of the materials for the reconstructive procedure. Reconstructive options include structural allograft transplantation, endoprosthetic replacement, and composite reconstruction with use of allografts and metal prostheses. As sarcomas are rare, it has been difficult to assemble a large number of cases. In 2011, Henderson et al. assembled a large series of endoprostheses and defined a classification scheme for failure41. This classification scheme described both mechanical and nonmechanical failures. Mechanical failures included soft-tissue failure (Type 1), aseptic loosening (Type 2), and structural failure (Type 3). Descriptions of each type of mechanical and nonmechanical failure were provided. Infection (Type-4 failure) is devastating and often results in removal of the implant. During reconstruction, leaving more bone and soft tissues around the resection can aid in the functional status postoperatively but unfortunately results in higher rates of local recurrence. The classification scheme described in this large multi-institutional series will also serve as a model for further collaborative efforts.
Giant Cell Tumor
Giant cell tumor of the spine and sacrum as well as recurrent giant cell tumors can pose a difficult surgical problem. Ruggieri et al., in a large series of patients with sacral giant cell tumors, highlighted the complications associated with the surgical treatment of these tumors42. In that series, adjuvants were not helpful with local control. While excellent local control was obtained, two deaths were associated with treatment. This series, like previous series, highlighted the need for better systemic therapy for the treatment of giant cell tumors. Thomas et al. reported promising results in a prospective trial targeting giant cells using a monoclonal antibody against RANKL43. The premise behind targeting RANKL is that receptors for RANKL exist on giant cells and activation of giant cells may rely in part on RANKL. It was hypothesized that if one were to block giant cell activation by binding RANKL, one could inhibit growth of giant cell tumors. This trial included thirty-seven patients with recurrent or unresectable giant cell tumors who underwent monthly subcutaneous injections of denosumab (Amgen, Thousand Oaks, California). The primary end point was tumor response, defined as elimination of at least 90% of giant cells or no radiographic progression of the target lesion up to Week 25. Thirty-five patients were included in the analysis, and thirty (86%) met the tumor-response criteria. Notably, the post-treatment histological findings demonstrated regions of new bone formation in places where histologically proven giant cell tumor had existed before treatment. It is unclear what risks are associated with long-term treatment with a RANKL inhibitor. At least one case of osteonecrosis of the jaw has been reported44.
Enthusiasm for targeting the RANK/RANKL pathway is leading others to consider its merits in the treatment of chondroblastoma and aneurysmal bone cysts, both of which possess giant cells. A Korean study evaluated the relative expression of RANK on giant cells from these tumors and compared it with RANK expression in giant cell tumors45. Chondroblastomas expressed higher levels of RANK than did giant cells from giant cell tumors, whereas aneurysmal bone cysts expressed lower levels of RANK than did giant cell tumors. Interestingly, the expression of osteoprotegerin was also much higher in chondroblastomas.
Bisphosphonates have also been used as an adjuvant for local control of giant cell tumors. A recent study evaluated the elution characteristics of zoledronic acid from acrylic bone cement and correlated them with their cytotoxic effects on stromal cells from a giant cell tumor in vitro46. The study demonstrated a dose-dependent effect on cellular cytotoxicity with incremental increases in zoledronic acid concentrations. Most of the zoledronic acid was released in the first twenty-four hours, and a plateau was reached by four days. The authors concluded that the cytotoxic effects of zoledronic acid were not mitigated by cement polymerization.
Klenke et al., in a large clinical series from the Mayo Clinic, evaluated the outcomes for 118 patients who underwent treatment of giant cell tumor of bone47. Although wide resection was associated with better local control, the authors were able to manage local recurrence well with additional surgery. Furthermore, they found a 4% risk of pulmonary metastasis, which was treated effectively with multimodal therapy. The authors found a lower risk of local recurrence in association with polymethylmethacrylate (PMMA) filling of bone defects than with allograft chips. They concluded that the vast majority of giant cell tumors of bone can be treated with curettage and PMMA packing, with a local recurrence rate of about 20%. A parallel study from the same center evaluated recurrent giant cell tumors and concluded that intralesional curettage and PMMA packing allowed for good local control even in the recurrent setting48. Furthermore, local failures were usually amenable to additional surgery.
Aneurysmal Bone Cyst
Additional work has been done to elucidate the underlying mechanisms behind aneurysmal bone cysts, which are known to express a TRE17/USP6 translocation. This translocation causes TRE17 (a ubiquitin-specific protease) to be highly expressed, whereas it is typically expressed at very low levels in human tissues. A recent study demonstrated that TRE17 expression leads to increased matrix metalloproteinase (MMP)-9 and increased MMP-10 activity via the NFκB pathway49. Both MMP-9 and 10 have been associated with degradation and remodeling of bone matrix. Furthermore, MMP-9 activity has been associated with release of VEGF, leading to increased vascularity. This study sheds new light on the etiology of aneurysmal bone cysts and may lead to subsequent molecular targets.
While we await systemic therapies to assist in the treatment of aneurysmal bone cysts, investigators are reporting the results of repetitive sclerotherapy in lieu of surgery. In the study by Varshney et al., ninety-four patients were randomized into two groups, with one group being managed with curettage and high-speed burring and the other group being managed with repetitive intraosseous sclerotherapy with use of polidocanol50. The authors hypothesized that the bone healing rate would be similar between groups. They reported a healing rate of 93% in the sclerotherapy group, compared with 85% in the curettage group. Furthermore, 31% of the patients in the sclerotherapy group had healing after one injection. The sclerotherapy procedure is an outpatient procedure, and pain relief was achieved immediately following the injection. The authors concluded that sclerotherapy is a safe and effective means of treating aneurysmal bone cysts.
The results of sclerotherapy compared favorably with those in another large series of aneurysmal bone cysts that were treated with curettage, high-speed burring, and argon beam coagulation51. In that series, local control was achieved in 92% of patients; however, the use of argon beam was associated with an increased fracture risk.
Benign Cartilaginous Tumors
Pedrini et al. studied 529 patients with multiple hereditary exostoses in order to correlate disease severity with various risk factors such as the presence of EXT1 as compared with EXT2 mutations, sex, and number of exostoses52. The authors found that female sex, having fewer than five sites with exostoses, the presence of EXT2 mutations, and the absence of EXT1/2 mutations were most commonly associated with a mild form of disease, as defined by the absence of deformity and functional limitations. On the other hand, male sex, EXT1 mutations, and having more than twenty sites with exostoses were associated with the severe form of the disease, as defined by the presence of deformities and functional limitations. Malignant degeneration occurred in 5% of the cohort, and the development of malignancy was not linked to EXT mutations, sex, the severity of the disease, or the number of sites with exostoses. Malignancy was more likely to occur in exostoses from the pelvis, scapula, and proximal part of the femur, and malignant degeneration occurred in patients with a family history of multiple hereditary exostoses (n = 20) rather than sporadic forms of the disease (n = 2).
EXT has been implicated in the pathogenesis of osteochondromas; however, its exact role has not been elucidated. One theory is that the development of an osteochondroma requires biallelic mutations in EXT genes. A recent study evaluated the presence of biallelic mutations in multiple hereditary osteochondromas, solitary osteochondromas, and chondrosarcomas53. There was no correlation between the presence of biallelic mutations in EXT and the pathogenesis of osteochondromas.
Additional work on multiple hereditary exostoses implicated microRNAs as a potential impediment to cartilage maturation and/or differentiation, potentially contributing to the pathogenesis and phenotypic variability in multiple hereditary exostoses54.
One study identified loss-of-function mutations in PTPN11 in individuals with metachondromatosis55. Metachondromatosis is a rare autosomal dominant condition in which patients manifest multiple enchondromas and multiple exostoses. Mutations were noted in eleven of seventeen families with known metachondromatosis, but there were no mutations in individuals with multiple enchondromatosis disorders, such as Ollier disease or Maffucci syndrome.