Osteosarcoma, or osteogenic sarcoma, is a rare disease but not a new one. With the advent of systemic chemotherapy in the 1960s, osteosarcoma was transformed from an essentially lethal disease to one in which 70% of patients with localized disease can now expect to be cured with complete resection and multiagent chemotherapy. In contrast, patients with unresectable, relapsed, or metastatic disease experience survival in the 20% range. For the estimated 30% who experience relapse after localized disease, survival remains poor. An update on the current understanding and clinical management of conventional high-grade osteosarcoma is provided.

Osteosarcoma most commonly occurs in adolescents and young adults. The disease peaks during the second decade of life during periods of rapid growth, with a second peak during the seventh and eighth decades of life. In the elderly, osteosarcoma often occurs as a secondary cancer, as a result of previous radiation therapy or an underlying bone disease such as Paget disease. As reported in a 2023 study, osteosarcoma accounts for 5% of all new pediatric cancers.¹ Osteosarcoma is diagnosed in approximately 450 children and young adults younger than 20 years. Osteosarcoma has a predilection for the metaphyseal portions of the long bones, with the distal femur and proximal tibia accounting for almost 50% of all osteosarcomas.²

The World Health Organization classifies osteosarcoma as either primary or secondary osteosarcoma. More than 90% of osteosarcomas are high grade. Primary osteosarcomas are further subdivided into intramedullary and
surface osteosarcomas.³ Conventional osteosarcoma is a primary intramedullary, high-grade malignant tumor in which neoplastic cells of mesenchymal origin produce osteoid (Figure 1). Conventional osteosarcoma has historically been classified into osteoblastic, chondroblastic, and fibroblastic histologic subtypes.

The understanding of the genomic landscape of osteosarcoma continues to evolve. Compared with other childhood cancers, osteosarcoma has an exceptionally chaotic genome. A number of somatic chromosomal lesions (eg, structural variations, copy number alterations) as well as inherited syndromes have been implicated in susceptibility to osteosarcoma development. The more common genetic diseases that predispose to osteosarcoma⁴ are listed in Table 1. Of these, Li-Fraumeni syndrome (germline p53 mutation) and hereditary retinoblastoma syndrome (germline Rb mutation) are among the best described in their consistent association with osteosarcoma pathogenesis.⁵ Both germline and mechanistic alterations in p53 have been found to be present in most osteosarcoma cases.^6,7 Other frequently described genetic alterations include amplifications of MDM2/4; RB mutations; and, less commonly, mutations in ATRX, cell cycle proteins, and the PTEN/PI3K pathway. Beyond this small set of mutations, the most common characterization of osteosarcoma is structural complexity via chromosomal rearrangements, copy number variation, kataegis, and chromothripsis.⁸

Recent genome sequencing efforts have further elucidated the highly diverse landscape of genetic aberrations with likely corresponding diverse biologic effects, and this is leading to emerging therapies targeting chromosomal instability, immune checkpoints, and antibody drug conjugates targeting membrane proteins overexpressed in osteosarcoma. Recent studies show that all are being explored as potential inroads to targeted therapies for osteosarcoma.^9,10 An integrated approach led by several collaborative initiatives will hopefully accelerate effective interventions.¹¹

Osteosarcoma can occur in any bone of the body. The most common sites of disease are the distal femur, the proximal tibia, and the proximal humerus. Tumors occurring in the appendicular skeleton outnumber those in the axial skeleton. Pain and swelling are the most common complaints. The initial reports of pain in the growing adolescent are often attributed to common benign conditions such as growing pains and trauma.² The median time from onset of symptoms to diagnosis is 4 months.

Although only 20% to 25% of all patients with a new diagnosis have radiographically detectable metastatic disease, all patients are presumed to have micrometastatic disease. This assumption is based on data from the prechemotherapy era, which showed that metastatic disease developed within 3 to 6 months after resection of the primary tumor in most patients with what was once considered nonmetastatic osteosarcoma on the basis of available imaging modalities.¹² The most common site of distant metastasis is the lungs, accounting for 80% of all metastases. The second most common site of metastatic spread is other bones.¹² Osteosarcoma, like all primary bone tumors, is staged according to the American Joint Committee on Cancer staging system. Noted changes in the eighth edition of the American Joint Committee on Cancer manual include the addition of specific primary tumor classifications for pelvis and spine tumors (T factor), incorporating criteria from those of the appendicular skeleton.

Few true, robust prognostic factors exist for clinicians to consider in osteosarcoma. The presence of clinically detectable metastases (stage of disease) remains the most powerful clinical factor in predicting prognosis in osteosarcoma.¹² As noted previously, whereas rate of survival for patients with nonmetastatic disease is approximately 70% with multiagent chemotherapy and wide surgical resection, long-term survival rate for patients with metastatic disease at diagnosis approaches 20%. Similarly, patients with recurrent or progressive disease have a rate of long-term survival of less than 20%.¹³ Other prognostic factors that have been used in osteosarcoma include patient age, tumor size, anatomic site of primary disease, lactate dehydrogenase level, and alkaline phosphatase level.^11,14 Thus far, none of these variables has demonstrated sufficient predictive value to serve as a basis for risk-adapted therapy stratification at diagnosis, such as the schemas used in the management of rhabdomyosarcoma. One predictor of outcome that has been used to stratify therapy in some studies is the percentage of tumor necrosis following induction chemotherapy. The percentage of necrosis is assessed histologically at the time of definitive resection, typically after 10 weeks of preoperative chemotherapy in most studies. Although grading has been shown to have a strong correlation with disease-free survival, it cannot be evaluated at initial diagnosis and thus cannot serve as a true prognostic factor to help in risk stratification of therapy.¹⁵ Despite this limitation, attempts have been made to modify therapy on the basis of necrosis assessment in an effort to improve survival. Unfortunately, these strategies have not proven to be of benefit. One explanation is that tumor necrosis in response to preoperative chemotherapy reflects innate tumor biology rather than chemotherapy effectiveness.¹⁶

Well-established guidelines exist for the initial evaluation of patients with suspected osteosarcoma. Before biopsy, plain radiography, MRI of the entire bone, imaging of the chest, and whole-body bone scan or 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) scan are recommended.

Plain radiography remains the preferred imaging method in the primary workup. Although radiography lacks many details such as identifying the extent of the soft-tissue involvement of the tumor, the appearance of conventional osteosarcoma is often pathognomonic. The radiographic appearance of osteosarcoma usually demonstrates a radiodense metaphyseal lesion with ill-defined margins. The normal trabecular pattern of the bone is distorted, and characteristic radiographic features such as the Codman triangle and sunburst appearance are common (Figure 2).

MRI demonstrates the extent of soft-tissue involvement, neurovascular proximity, the extent of bone marrow involvement, and the presence of skip lesions (Figure 3). The whole length of the bone needs to be visualized in at least one plane. Axial cuts are useful to delineate the relationship of the tumor to the neurovascular structures and hence are most useful in determining limb salvage resectability (Figure 4). A number of functional MRI sequences have been described to assess treatment response. Diffusion-weighted imaging, dynamic contrast-enhanced perfusion imaging, and magnetic resonance spectroscopy have expanded the role of MRI to include lesion characterization, treatment response, and the detection of postsurgical recurrence.¹⁷

Chest CT is more sensitive than chest radiographs in detecting lung metastasis smaller than 1 cm. Studies have demonstrated that nodules that are small (smaller than 5 mm), solitary, or unchanged on serial radiologic studies are typically benign. Peripheral lesions larger than 5 mm are likely to be metastatic (Figure 5). Lesions that do not fit into these criteria should still be followed with serial CT examinations, and any increase in size should prompt a biopsy to clarify the diagnosis. Although the CT scan is the most sensitive tool available, CT scanning of the chest has been shown to underestimate the number of histologically proven osteosarcoma nodules.¹⁸

Either nuclear medicine imaging in the form of technetium Tc-99 whole-body bone scan (Figure 6) or 18F-FDG PET scan (Figure 7) can be used to evaluate for distant bone metastases. PET has the additional advantage of identifying nonbony sites of disease and quantifying treatment response. Several studies have compared bone scan with PET scan with conflicting results.^19,20 A meta-analysis concluded that PET scan for osteosarcoma staging showed high sensitivity and specificity.²¹ Recent recommendations from the Children’s Oncology Group in a 2023 report cite PET as strongly recommended for staging in both Ewing sarcoma and osteosarcoma.²² Whole-body MRI is a promising emerging modality that provides high tissue contrast and does not use ionizing radiation, but has not been validated in osteosarcoma to the same extent as PET. 18F-FDG PET-MRI combines the metabolic imaging of 18F-FDG PET with the advantages of whole-body MRI and is an alternative where available.

Multiagent chemotherapy and surgery are the standard approach to the treatment of patients with osteosarcoma. Historical series demonstrated that most patients who undergo only surgery for management of osteosarcoma experience pulmonary metastases within 6 months. Multiagent chemotherapy regimens pioneered in the 1970s markedly improved survival in osteosarcoma. Survival outcomes, however, have remained unchanged in the past 30 years. Currently, all modalities used in the management of malignancy are being brought to bear in the management of this disease, including surgery, chemotherapy, radiation therapy, immunotherapy, gene therapy, and therapies aimed at the tumor microenvironment.