Ewing sarcoma was first described in the medical literature in 1921 by Dr. James Ewing. Since that time, the diagnostic requirements, treatment, and survival have dramatically evolved. There are four categories of small round cell sarcomas, and these tumors share morphologic features but have distinct molecular signatures and clinical behavior. Ewing sarcoma is the most common of the small round cell sarcomas and is defined by the pathognomonic FET-ETS gene fusions. It occurs equally in the central and peripheral skeleton and the same can be said for both long and flat bones. Approximately 10% to 20% of Ewing sarcomas arise in soft tissue. Both the bone and soft-tissue subtypes have identical molecular and immunohistochemical profiles.

Patients with Ewing sarcoma are treated with a combination of chemotherapy, radiation therapy, and surgery. Multiagent chemotherapy usually consists of vincristine, doxorubicin, and cyclophosphamide, alternating with ifosfamide and etoposide (VDC/IE) every 2 weeks. Induction chemotherapy is administered before surgical resection. Surgery is followed by adjuvant chemotherapy. Radiation therapy can be used for definitive local control. The decision to use definitive radiation is a complex multidisciplinary process, but is often reserved for tumors that are unresectable.

In 2020, the World Health Organization Classification of Soft Tissue and Bone Tumours recognized four categories of small round cell sarcomas: Ewing sarcoma, CIC-rearranged sarcoma, sarcoma with BCOR alterations, and round cell sarcoma with EWSR1-non-ETS fusions.^1,2 Ewing sarcoma is the second most common primary malignancy of bone in children and adolescents. Surveillance, Epidemiology, and End Results data confirm the annual incidence in children and adolescents to be 3 of 1,000,000 cases per year, with approximately 64% presenting with localized and/or regional disease, 27%
with metastatic disease, and 9% unstaged. The peak incidence occurs in the second decade of life with a slight male predominance.³ Ewing sarcoma is nine times more likely to develop in US Caucasians than in African Americans and shows a decreased incidence in Asian populations, indicating a disparate incidence among certain ethnicities.⁴ Extraosseous Ewing sarcoma occurs less frequently, in approximately 4 of 1 million cases per year.^5,6

The most common osseous sites of presentation are the lower extremities and pelvis. Within the long bones, tumors typically occur within the diaphyseal or metadiaphyseal region. Extraosseous Ewing sarcoma more commonly occurs in the truncal region and extremities.

The earliest symptom of Ewing sarcoma is progressive pain. As the tumor increases in size, it can lead to visible or palpable swelling of the affected area. The pelvis and spine can mask tumor bulk and may not be as easily discernible as extremity tumors. Because of this, Ewing sarcoma of the pelvis can be quite locally advanced at presentation. Other symptoms can include fever, decreased appetite, weight loss, and fatigue. Laboratory testing may show anemia as well as an inflammatory response with increases in C-reactive protein level, erythrocyte sedimentation rate, lactate dehydrogenase level, and leukocytosis.⁷

Physical examination frequently reveals a firm mass, or asymmetric fullness of the affected extremity or anatomic compartment. Erythema may be present, particularly when a rapidly growing mass is attenuating the overlying skin. Spine-based presentations can cause nerve or cord compression, resulting in bowel or bladder dysfunction, radiculopathy, or myelopathy. Regional lymph node involvement is uncommon but arises more frequently in extraosseous tumors.⁸

Initial workup of Ewing sarcoma will include a history and physical examination along with plain radiography. Ewing sarcoma typically presents as a permeative, lytic lesion within the diaphysis or metadiaphysis of long bones (Figure 1). Reactive bone formation such as periosteal reaction (eg, onion skin or Codman triangle) or marginal sclerosis can also be seen. MRI with and without contrast will evaluate soft-tissue extension and proximity to critical structures. MRI of the entire length of the bone screens for skip metastasis. Ewing sarcoma typically appears isointense to muscle on T1-weighted sequences and is hyperintense on T2-weighted sequences (Figure 2). Intense contrast enhancement is typical and can be either homogenous or heterogeneous depending on the degree of necrosis and other factors. Adjacent perilesional edema is often present.

Following initial imaging workup, biopsy should be performed to confirm the diagnosis. When possible, the biopsy should be performed at a sarcoma treatment center. Biopsy planning is a coordinated effort between the radiologist and treating surgeon to ensure that the biopsy tract will be in line with the ultimate surgical resection incision and avoids contamination of uninvolved muscle compartments, joint spaces, or neurovascular bundles. The three categories of biopsy techniques for bone cancer are core needle biopsy, fine-needle aspiration, and open biopsy. Core needle biopsy is recommended because it will provide an adequate amount of tissue to perform genomic analysis compared with fine-needle aspiration. Core needle biopsy also has the benefit of lower rates of surrounding tissue contamination compared with open biopsy.⁹ In addition to a thorough histologic analysis, ancillary testing including immunohistochemistry, fluorescence in situ hybridization, and genomic sequencing is critical for differentiating Ewing sarcoma from other small round cell tumors. Immunohistochemistry should demonstrate strong and diffuse CD99 staining. Fluorescence in situ hybridization should identify an EWSR1 rearrangement, although a negative result does not exclude Ewing sarcoma because other fusions have been described (ie, FUS). RNA sequencing is highly sensitive and specific and may be required in difficult cases.

Following a biopsy-proven diagnosis of Ewing sarcoma, staging studies are performed. Consensus guidelines recommend contrast MRI of the primary site, chest CT, 18F fluorodeoxyglucose-positron emission tomography (PET)-CT, consideration of bone marrow biopsy and/or MRI of the spine and pelvis with and without contrast, cytogenetics and/or molecular studies, and assessment of lactate dehydrogenase level. Staging assessment has important implications for both prognosis and treatment. Metastatic status is the most important prognostic indicator in Ewing sarcoma as survival rates decrease significantly for those with metastatic disease compared with those presenting with localized disease.¹⁰ The key elements in assigning stage for Ewing sarcoma are the presence or absence of metastasis, size, and extent of local spread. Chest CT and head-to-toe 18F fluorodeoxyglucose-PET-CT are performed to evaluate for metastatic disease, with the lung being the most common site followed by bone. Chest CT is a sensitive imaging test for pulmonary metastasis, but often fails to distinguish reactive subcentimeter pulmonary nodules from early metastatic lesions. A meta-analysis of total body fluorodeoxyglucose PET demonstrated a high degree of sensitivity (96% confidence interval, 91% to 99%) and specificity (92% confidence interval, 87% to 96%) in assigning stage for Ewing sarcoma.¹¹ Bone marrow biopsy and/or MRI of the spine and pelvis with and without contrast should be considered.¹⁰ Bone marrow assessment is performed to determine whether microscopic evidence of marrow metastasis exists. In patients presenting without evidence of metastatic disease, the incidence of having a positive bone marrow biopsy has been reported to be as low as 1.2%. Because of the sensitivity of PET-CT, the physician could consider omitting bone marrow biopsy in patients with only localized disease because PET-CT has been shown in a 2021 study to demonstrate 100% sensitivity and 96% specificity in detecting bone marrow metastasis.¹² Patients presenting with marrow involvement have a poor prognosis.¹³

Laboratory evaluation is useful for determining the health status of the patient at presentation and has bearing on prognosis. Complete blood cell count, comprehensive metabolic panel, erythrocyte sedimentation rate, C-reactive protein level, and lactate dehydrogenase level are recommended. Erythrocyte sedimentation rate and C-reactive protein level are nonspecific and are often elevated in patients with Ewing sarcoma, particularly if they exhibit fever, malaise, or decreased appetite. Microcytic anemia may be present, indicating a chronic marrow response to tumor cytokine production. Normal complete blood count and platelet count do not preclude marrow infiltration. An elevated lactate dehydrogenase level at the time of diagnosis is associated with an increased risk of disease recurrence and decreased survival.^14,15,16

Gross examination of Ewing sarcoma that is untreated will show gray-white tissue that will contain hemorrhagic areas as well as necrosis. Histologic examination of Ewing sarcoma reveals a small round cell tumor (Figure 3). In most cases, the cells are uniform and contain round nuclei with finely stippled chromatin and subtle nucleoli. There are cases with larger tumor cells with more conspicuous nucleoli.¹⁷ In approximately one-half of tumors, glycogen deposition within the cytoplasm causes positive periodic acid-Schiff staining. Based on morphology alone, Ewing sarcoma closely resembles other neoplastic conditions such as hematologic malignancies, histiocytic disorders, small cell osteosarcoma, mesenchymal chondrosarcoma, metastatic neuroblastoma, rhabdomyosarcoma, monophasic synovial sarcoma, and chronic infection.¹⁸

Immunohistochemistry provides important specificity to routine morphologic observations. CD99 is a sensitive marker for Ewing sarcoma, with strong expression in 95% of tumors.¹⁹ However, CD99 is not specific for Ewing sarcoma, with expression occurring throughout a variety of tumor types as well as healthy tissue. More specific immunohistochemical markers for Ewing sarcoma include Friend leukemia virus integration 1 (FLI1); however, approximately 15% of tumors do not contain the FLI1 translocation.²⁰ Ewing sarcoma is definitively diagnosed by the pathognomonic chromosomal translation that fuses a member of the FET family of RNA-binding proteins with members of the family of ETS transcription factors.²¹ A 2020 study showed that the EWSR1-FLI1 is the most common fusion (73%), followed by EWSR1/FUS-ERG (15%), EWSR1/FUS-FEV (5%), and EWSR1-ETV1/4 (1%).²¹ The EWSR1-FLI1 is classified based on the t(11;22)(q24;12) chromosomal translocation, which joins exon 7 of EWSR1 to exon 6 (type I; 60% of cases) versus exon 5 (type II; 25% of cases) of FLI1.²² It was initially thought that the type I fusion may have a prognostic effect; however, this has not been shown in any prospective trials.^23,24

Before the introduction of systemic chemotherapy for Ewing sarcoma, the 5-year survival rate was 10%.²⁵ In the early 1970s, chemotherapy was introduced and resulted in significant improvement in outcomes. In the late 1980s, the addition of ifosfamide and etoposide alternating with doxorubicin, vincristine, cyclophosphamide, and dactinomycin was found to improve 5-year event-free survival to 69% in those who presented with localized disease.²⁶ Intensifying treatment cycles of VDC/IE from every 3 weeks to every 2 weeks further improved event-free survival in those with localized disease to 73%.²⁷ In the United States, the standard treatment is vincristine, doxorubicin, and cyclophosphamide, alternating every 2 weeks with ifosfamide and etoposide. Induction chemotherapy is administered for a total of six cycles, which is then followed by local therapy. Following local therapy, an additional 8 cycles of chemotherapy are administered, for a total of 14 cycles.

Prognosis for Ewing sarcoma is affected by tumor location, size, patient age, and presence of metastasis. Approximately 20% to 30% of patients will present with metastatic disease and this is the strongest predictor of poor prognosis.²⁸ Adult patients and those with tumors larger than 8 cm or tumors located in the pelvis have worse survival.^29,30 There have been multiple studies that have demonstrated that patients with bone or bone marrow metastasis have poorer survival rates compared with those with lung metastasis.^10,29 Patients with lung-only metastasis have experienced improved survival with whole-lung radiation therapy.³¹ There are also data to suggest that those presenting with localized disease have worse survival when local control occurs later than 16 weeks following the induction of chemotherapy.³² Response to chemotherapy has a prognostic effect on local recurrence. Patients with greater than 90% tumor necrosis were found to have an 86% local recurrence-free survival compared with 51% in patients with less than 90% tumor necrosis.³³ The best responders, however, are patients with 100% tumor necrosis. This group of patients has been shown to have superior overall survival, local recurrence-free survival, and metastasis-free survival compared with patients with 90% to 99% necrosis.³⁴ As discussed in a 2021 study, detectable circulating tumor DNA at diagnosis and persisting following neoadjuvant treatment has been shown to be associated with inferior survival.³⁵