Osteosarcoma Variant Imaging
- Author: Geoff Hide, MBBS, MRCP, FRCR; Chief Editor: Felix S Chew, MD, MBA, MEd more...
Osteosarcoma is the most common primary malignant tumor of bone, excluding plasma cell myeloma. Approximately 75% of all osteosarcomas are of the classic or conventional type, and the remaining 25% comprise the osteosarcoma variants, which are the subject of this article. The variants are a heterogeneous group of osteosarcomas with a range of different imaging and behavioral features.[1, 2, 3, 4, 5, 6]
Preferred modalities for evaluating primary disease are radiography, MRI, and sometimes computed tomography (CT) scanning. Staging is always performed by using chest CT scanning to detect pulmonary metastases. Isotopic bone scanning is generally used to detect skeletal metastases or synchronous tumors, but whole-body MRI may replace this study.[7, 8, 9, 10]
The overall prognosis for patients with osteosarcoma depends on the stage of the tumor at presentation. Without metastases, long-term survival is in the order of 60-85%.
Telangiectatic osteosarcoma (as shown in the images below) has been considered more aggressive than classic osteosarcoma, but studies of long-term survival after optimum treatment now indicate that the aggressiveness of telangiectatic osteosarcoma is similar to that of the classic type.
Intraosseous low-grade osteosarcoma generally has a good prognosis. Gnathic osteosarcoma (as shown in the images below) is less frequently associated with metastatic spread than is conventional osteosarcoma, but local disease recurrence is often problematic.
The prognosis for intracortical osteosarcoma is unclear because of its rarity. Both small-cell and secondary osteosarcoma are generally associated with a poor prognosis.[11, 2] High-grade surface osteosarcoma has a prognosis similar to that for a conventional osteosarcoma, and the prognosis for periosteal osteosarcoma (shown in the images below) is better than that for conventional osteosarcoma.
The prognosis for multicentric osteosarcoma is dire.
Conventional osteosarcoma is most frequent in areas of high skeletal growth, especially the metaphyseal regions of the distal femur, proximal tibia, and proximal humerus. Most osteosarcoma variants follow a similar distribution, with the exception of gnathic (mandible and maxilla) lesions, intracortical lesions (rare but more typically diaphyseal), periosteal lesions (more typically diaphyseal), and secondary osteosarcomas.[4, 5, 12] The last osteosarcomas frequently occur in the pelvis and proximal femur, often in association with Paget disease (as demonstrated in the images below).
Telangiectatic osteosarcoma is generally lytic, with a periosteal reaction and soft-tissue mass. When the tumor margins are well defined, it may mimic an aneurysmal bone cyst. Small-cell osteosarcoma appears similar to a conventional osteosarcoma; it often has mixed areas of sclerosis and lysis. Intraosseous low-grade osteosarcoma may be lytic, sclerotic, or mixed; it often has deceptively benign features of well-defined margins and the absence of periosteal changes or a soft-tissue mass.
Gnathic tumors may be lytic, sclerotic, or mixed, and bone destruction, periosteal reaction, and soft-tissue extension are common. Intracortical osteosarcomas are described as radiolucent and geographic, and they contain a small amount of internal mineralization. High-grade surface osteosarcomas are shown as broad-based soft-tissue masses with varying degrees of mineralization arising from the surface of the bone.
Parosteal osteosarcomas are typically densely ossified tumors arising from a broad base on the adjacent bone. Unlike osteochondromas, parosteal osteosarcomas involve no continuation of the medullary cavity into the tumor.
CT scanning is helpful in the evaluation of a variety of the osteosarcoma variants. It may demonstrate fluid levels in telangiectatic osteosarcoma, and a contrast-enhanced CT scan can be helpful in discriminating such a lesion from an aneurysmal bone cyst. Telangiectatic osteosarcoma differs from an aneurysmal bone cyst in that the former has a rim of tumor cells that surrounds the cystic spaces. This tissue rim shows typically nodular enhancement after the intravenous administration of contrast material.
CT scanning is useful in the evaluation of bone changes occurring in areas of complex anatomy. Examples are the changes in the maxilla or mandible that are associated with gnathic osteosarcoma and those in the pelvis that are associated with secondary osteosarcoma. CT scanning provides useful information about the surface osteosarcoma variants, including parosteal, periosteal, and surface high-grade tumors.
When appropriate and performed in consultation with an orthopedic oncologist, CT scanning can be useful in guiding biopsy.
Magnetic Resonance Imaging
MRI is the optimum technique for local staging of osteosarcomas. In certain cases, MRI is combined with CT scanning. MRI accurately demonstrates the extent of a tumor within bone and soft tissue.
At least 1 sequence, either a T1-weighted or a short-tau inversion recovery (STIR) sequence, should be performed to image the entire bone. This is necessary to exclude skip lesions that are present within the same bone but are distant from the primary lesion. Periosteal osteosarcoma is typically a chondroblastic lesion, and the tumor usually has high signal intensity on T2-weighted MRIs.[13, 6]
MRI is more sensitive than CT scanning in demonstrating fluid-fluid levels in telangiectatic osteosarcoma because of its greater intrinsic soft-tissue contrast. Fluid-fluid levels can be seen in benign bone lesions as well, particularly aneurysmal bone cysts.
Histologic confirmation of the nature of the tumor is initially required ; the analysis should be performed after MRI and in consultation with the tumor surgeon. Biopsy must be performed after the MRI study because hemorrhage occurring at the time of biopsy alters the signal intensity characteristics of the tumor at subsequent MRI examinations. The site of the biopsy track must be planned to prevent contaminating the muscle compartments that the surgeon would not otherwise excise. The biopsy track is removed during surgery, and consideration should be given to marking the track with suture material or dye if there will be a delay between biopsy and formal excision.
Assessing treatment response
Oka et al evaluated whether the average apparent diffusion coefficient (ADC) or the minimum ADC provides a better assessment of patient response to chemotherapeutic osteosarcoma treatment.
Diffusion-weighted and magnetic resonance imaging (MRI) scans were performed on 22 patients with osteosarcoma, before and after chemotherapy, using the average and minimum ADCs. The authors found that in patients who responded well to chemotherapy, the minimum ADC ratio (using the prechemotherapy and postchemotherapy scan results) was significantly higher than it was in patients who responded poorly to treatment. However, the average ADC ratio was not significantly different between good and poor responders.
The authors concluded that the minimum ADC is a better tool than the average ADC for evaluating the chemotherapeutic response of patients with osteosarcoma.
Ultrasonography can demonstrate the soft-tissue extent of the tumor, but it cannot be used to evaluate the intramedullary component of the lesion. Ultrasonography is not routinely used in staging such lesions. Sonography can be useful in guiding percutaneous biopsy of the soft-tissue component of the tumor, again in consultation with an orthopedic oncologist.
Osteosarcomas typically show increased uptake of radioisotope; this characteristic makes bone scans sensitive but not specific. Bone scans are most useful in excluding multifocal disease. Multiple-gated acquisition (MUGA) cardiac scans may be required to monitor the toxic effects of certain chemotherapeutic agents.
In a retrospective study, Kaste et al estimated the likelihood of developing second cancers and of related mortality in pediatric patients undergoing thallium-201 (201Tl) bone imaging for osteosarcoma. The study's 73 patients each underwent three 201Tl studies, receiving a median dose of 4.4 mCi (162.8 MBq) (range, 2.2-8.4 mCi [81.4-310.8 MBq]) per study. Males received a total median cumulative radiation dose of 18.6 rem (186 mSv) (range, 8.4-44.2 rem [84-442 mSv]), and females received a total of 21.5 rem (215 mSv) (range, 7.0-43.8 rem [70-438 mSv]).
The authors estimated that the incidence of excess cancers was as follows:
Exposure to 201Tl imaging by age 5 years: 6.0 cancers per 100 males; 13.0 cancers per 100 females
Exposure by age 15 years: 2.0 cancers per 100 males; 3.1 cancers per 100 females
The estimated mortality resulting from these excess cancers was as follows:
Exposure to 201 Tl by age 5 years: 3.0 deaths per 100 males; 5.2 deaths per 100 females
Exposure by age 15 years: 1.0 deaths per 100 for males; 1.4 deaths per 100 for females
The authors concluded that reduction of 201Tl exposure will be necessary before thallium becomes a viable means of imaging osteosarcoma in younger patients.
Murphey MD, Robbin MR, McRae GA, et al. The many faces of osteosarcoma. Radiographics. 1997 Sep-Oct. 17(5):1205-31. [Medline].
Puranik AD, Purandare NC, Bal MM, Shah S, Agrawal A, Rangarajan V. Extraskeletal osteosarcoma: An uncommon variant with rare metastatic sites detected with FDG PET/CT. Indian J Med Paediatr Oncol. 2014 Jan. 35 (1):96-8. [Medline].
Hewitt KM, Ellis G, Wiggins R, Bentz BG. Parosteal osteosarcoma: case report and review of the literature. Head Neck. 2008 Jan. 30(1):122-6. [Medline].
Chow LT, Wong SK. Epiphyseal osteosarcoma revisited: four illustrative cases with unusual histopathology and literature review. APMIS. 2015 Jan. 123 (1):9-17. [Medline].
Chakravarthi PS, Kattimani VS, Prasad LK, Satish PR. Juxtacortical osteosarcoma of the mandible: Challenges in diagnosis and management. Natl J Maxillofac Surg. 2015 Jan-Jun. 6 (1):127-31. [Medline].
Almeida E, Mascarenhas BA, Cerqueira A, Medrado AR. Chondroblastic osteosarcoma. J Oral Maxillofac Pathol. 2014 Sep-Dec. 18 (3):464-8. [Medline].
Mirra JM. Osseous tumours of intramedullary origin. Mirra JM, ed. Bone Tumours: Clinical, Radiologic and Pathologic Correlations. Philadelphia, Pa: Lea & Febiger; 1989. 248-438.
Panuel M, Gentet JC, Scheiner C, et al. Physeal and epiphyseal extent of primary malignant bone tumors in childhood. Correlation of preoperative MRI and the pathologic examination. Pediatr Radiol. 1993. 23(6):421-4. [Medline].
Parham DM, Pratt CB, Parvey LS, et al. Childhood multifocal osteosarcoma. Clinicopathologic and radiologic correlates. Cancer. 1985 Jun 1. 55(11):2653-8. [Medline].
Resnik D, Kyriakos M, Greenaway GD. Tumors and tumor-like lesions of bone: imaging and pathology of specific lesions. Diagnosis of Bone and Joint Disorders. 4th ed. Philadelphia, Pa: WB Saunders Co; 2002. vol 4: 3800-33.
Yang JY, Kim JM. Small cell extraskeletal osteosarcoma. Orthopedics. 2009 Mar. 32(3):217. [Medline].
Barker JP, Monument MJ, Jones KB, Putnam AR, Randall RL. Secondary osteosarcoma: is there a predilection for the chondroblastic subtype?. Orthopedics. 2015 May. 38 (5):e359-66. [Medline].
Fox C, Husain ZS, Shah MB, Lucas DR, Saleh HA. Chondroblastic osteosarcoma of the cuboid: a literature review and report of a rare case. J Foot Ankle Surg. 2009 May-Jun. 48(3):388-93. [Medline].
Davicioni E, Wai DH, Anderson MJ. Diagnostic and prognostic sarcoma signatures. Mol Diagn Ther. 2008. 12(6):359-74. [Medline].
Oka K, Yakushiji T, Sato H, et al. The value of diffusion-weighted imaging for monitoring the chemotherapeutic response of osteosarcoma: a comparison between average apparent diffusion coefficient and minimum apparent diffusion coefficient. Skeletal Radiol. 2009 Nov 19. [Medline].
Kaste SC, Waszilycsak GL, McCarville MB, et al. Estimation of potential excess cancer incidence in pediatric 201Tl imaging. AJR Am J Roentgenol. 2010 Jan. 194(1):245-9. [Medline].