INTRODUCTION — Ewing sarcoma (EWS) is a rare malignancy that most often presents as an undifferentiated primary bone tumor; less commonly, it arises in soft tissue (extraosseous Ewing sarcoma). Several terms were previously used to describe this entity, including Ewing sarcoma family of tumors (EFT), peripheral primitive neuroectodermal tumor (PNET; previously called peripheral neuroepithelioma), and Askin tumor of the chest wall [1]. Because these tumors share similar histological and immunohistochemical characteristics and unique, non-random chromosomal translocations, they are considered to have a common origin [2-5]. The pathology classification system from the World Health Organization refers to these tumors as Ewing sarcoma (EWS), and that nomenclature will be used here.

EWS has a peak incidence between the ages of 10 and 20 years (70 percent of affected patients are under the age of 20), a tendency towards early dissemination to the lungs, bone, and bone marrow, and responsiveness to chemotherapy and radiation therapy (RT). Advances in multidisciplinary management over the past 30 years have resulted in a marked improvement in long-term survival. In data derived from the Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Institute, five-year survival rates for patients with Ewing sarcoma rose from 36 to 56 percent during the periods 1975 to 1984 and 1985 to 1994 [6]. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)

This topic will discuss the management of EWS. The epidemiology, pathology, molecular genetics, clinical presentation, and diagnosis of these tumors, surgical principles, indications for limb-sparing surgery, and indications for RT are discussed separately. In addition, central nervous system (supratentorial) PNET tumors and undifferentiated sarcomas lacking classic Ewing sarcoma chromosomal translocations (so-called Ewing-like sarcomas) are biologically distinct from Ewing sarcoma. Management of these tumors is discussed separately.

●(See "Epidemiology, pathology, and molecular genetics of the Ewing sarcoma family of tumors".)

●(See "Clinical presentation, staging, and prognostic factors of the Ewing sarcoma family of tumors".)

●(See "Uncommon brain tumors", section on 'Ewing sarcoma/primitive neuroectodermal tumor'.)

●(See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)

GENERAL TREATMENT PRINCIPLES — Despite the fact that fewer than 25 percent of patients have overt metastases at the time of diagnosis, EWS is considered a systemic disease. Because of the high relapse rate (80 to 90 percent) in patients undergoing local therapy alone, it is surmised that the majority of patients have subclinical metastatic disease at the time of diagnosis, even in the absence of overt metastases.

Chemotherapy can successfully eradicate these deposits, and modern treatment plans all include chemotherapy, usually administered prior to and following local treatment. For patients with localized disease, the addition of intensive multiagent chemotherapy to local therapy has had a dramatic impact on survival, and reported five-year survival rates are now approximately 70 percent [7-16].

A subset of patients with advanced disease may be cured by multimodality therapy, although the long-term survival rates are clearly lower than for patients with localized disease [17]. Approaching a patient who has metastatic EWS with noncurative intent is rarely, if ever, appropriate since it is not possible to predict a priori which patients will be cured. Because of these issues, clinicians experienced in the treatment of EWS must direct the surgery and radiation therapy (RT), and coordination with the medical/pediatric oncologist is essential [11,18,19].

Adult patients — Treatment of adults with EWS should be guided by the same general principles as are used for younger individuals.

Multiple studies have consistently demonstrated that older age is an adverse prognostic factor in EWS. Some studies have defined "older" patients as being those over the age of 15 at initial diagnosis, while others have used a cutpoint of 18 years. It is not clear if the worse prognosis in older individuals is due to biological differences or differences in treatment regimens (ie, regimens used in pediatric versus those used in adult medical oncology). Adults who are treated with modern adjuvant and neoadjuvant chemotherapy for localized EWS may do as well as children [20-22].

Fewer than 5 percent of cases of EWS arise in adults over the age of 40. There are no clinical trials that address treatment in adults; the vast majority of published studies have specifically excluded older individuals. The available information on treatment and prognosis is limited to seven single-institution series of patients over the age of 15 who were treated for EWS, and a retrospective analysis of data on patients with Ewing sarcoma reported to the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute, which included 383 patients diagnosed at age 40 or older [20,22-30]. The largest retrospective series of 102 patients treated between 1977 and 2005 showed an overall survival of 69 percent and an event-free survival of 52 percent [28]. A subsequent cohort (treated between 1993 and 2007) had better outcomes (five-year overall and event-free survival of 73 and 60 percent, respectively). This cohort routinely had the addition of ifosfamide and etoposide, similar to pediatric treatment protocols, and the authors concluded that their survival rates were similar to those of children.

TREATMENT FOR LOCALIZED DISEASE

Chemotherapy — Most modern treatment plans utilize initial (induction or neoadjuvant) chemotherapy followed by local treatment and additional chemotherapy. Reduction of local tumor volume is accomplished in the majority of patients, and this can facilitate resection. This is particularly important with regard to limb-sparing procedures for extremity lesions, but it may also be important for rib, chest wall, and vertebral primaries [31-33]. Initially, chemotherapy was used in the adjuvant setting to control metastatic disease, but it is now administered prior to local therapy (neoadjuvant therapy) to improve local control as well [34]. Since most treatment failures are attributable to systemic metastatic disease, local therapy considerations should never compromise the administration of effective systemic therapy. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)

Although patients with extraosseous Ewing sarcoma (EES) have been treated in the past on protocols for rhabdomyosarcoma and have a similar response to multimodality therapy [13], EES responds to the same chemotherapy regimens as osseous Ewing sarcoma, and patients with these tumors should be treated similarly [35-38]. More recent data suggest that patients with EES have slightly superior outcomes compared with patients with osseous Ewing sarcoma [39].

Chemotherapy treatment has evolved, largely due to the efforts of several cooperative groups [34]:

●In the first Intergroup Ewing sarcoma study (IESS-I), the combination of vincristine, doxorubicin, cyclophosphamide, and dactinomycin (VDCA or VACA) was associated with a significantly better five-year relapse-free survival than vincristine, dactinomycin, and cyclophosphamide (VAC) alone or VAC plus adjuvant bilateral pulmonary irradiation (60 versus 24 versus 44 percent, respectively) [9].

●Increasing the doxorubicin dose intensity during the early months of therapy further improved response, and in the second Intergroup study (IESS-II), the five-year relapse-free survival rates using intermittent high-dose four-drug therapy improved to 73 percent for non-pelvic lesions [40]. Because of concerns about limiting the dose intensity of doxorubicin in regimens containing dactinomycin [41], dactinomycin was omitted from most trials thereafter, with no adverse impact on long-term outcome.

●Adding alternating cycles of ifosfamide and etoposide (I/E) to a vincristine, doxorubicin, and cyclophosphamide (VDC) backbone (VDC/IE) provides further benefit [14,42-44]. In a randomized phase III clinical trial (the IESS-III study), the addition of I/E to VDCA was associated with significantly better five-year relapse-free survival compared with VDCA alone (69 versus 54 percent, respectively) in patients with localized (nonmetastatic) disease but not in those with metastatic disease [42].

As a result of these data, standard chemotherapy for EWS in the United States includes VDC/IE [42]. Typically, four to six cycles of chemotherapy are given before local therapy in the absence of disease progression followed by local treatment, and then additional cycles of the same treatments are given postoperatively for a total of 14 to 17 cycles. Relief of pain, decrease in tumor size, fall in lactate dehydrogenase (LDH) level, radiologic improvement, and evidence of necrosis in the resected specimen are all evidence of the antitumor effect of the prescribed regimen.

Dose-intense chemotherapy — The sensitivity of EWS to alkylating agents, which have a steep dose-response curve, has prompted the evaluation of dose-intense regimens. In a randomized phase III clinical trial conducted by the Children's Oncology Group (COG), dose escalation without hematopoietic cell support did not improve outcomes in patients with newly diagnosed localized disease [45]. Furthermore, concerns for an increased risk of secondary malignancies in patients receiving dose-intense therapy have tempered enthusiasm for this approach.

A different approach to intensifying therapy is with interval compression, in which chemotherapy cycles are administered every 14 days instead of every 21 days [46]. The results of a randomized phase III trial by COG established interval-compressed therapy with alternating cycles of VDC/IE as the preferred North American regimen for front-line treatment of children with localized EWS (table 1). In this study, 587 patients with localized EWS were randomly assigned to receive VDC/IE every 21 or 14 days. This trial demonstrated a significant event-free survival (EFS) benefit for interval-compressed VDC/IE given every 14 days, as compared with the same chemotherapy given at 21-day intervals (five-year EFS 73 versus 65 percent) [47]. The toxicity of both regimens was similar.

Data from another trial also supports the use of interval-compressed VDC/IE over other regimens. In preliminary results of a randomized European trial (EE2012) of 640 patients with EWS, initial treatment with interval-compressed VDC/IE improved EFS and overall survival compared with the vincristine, ifosfamide, doxorubicin, and etoposide (VIDE) regimen [48]. However, combining this regimen with chemotherapy effective in relapsed disease (eg, topotecan/cyclophosphamide) does not confer additional clinical benefit. In a separate randomized phase III COG trial (AEWS1031) conducted in 629 patients with localized treatment-naïve EWS, the addition of vincristine/topotecan/cyclophosphamide to the interval-compressed VDC/IE regimen did not improve EFS or overall survival [49,50].

The role of consolidative dose-intense chemotherapy followed by autologous hematopoietic cell support for high-risk localized disease was directly addressed in the European Ewing Tumour Working Initiative of National Groups (Euro-EWING) 99/EWING 2008 trial, which randomly assigned 214 patients after receiving six courses of VIDE. Patients were eligible for randomization if they were younger than 50 years with a poor histologic response (76 percent of the cohort), or had a tumor volume at diagnosis ≥200 mL if unresected, initially resected, or resected after radiation therapy (RT) [51]. The randomization was to a single course of busulfan plus melphalan followed by autologous hematopoietic cell rescue versus seven courses of standard chemotherapy (vincristine, dactinomycin, and ifosfamide [VAI]). At a median follow-up of 7.8 years, eight-year EFS was significantly better in the high-dose chemotherapy group (61 versus 47 percent), and overall survival also favored high-dose chemotherapy (65 versus 56 percent). Severe acute toxicities were more common in the high-dose chemotherapy group, and three patients died in this group, two of treatment-related toxicity; the third patient did not receive dose-intense chemotherapy or high-dose chemotherapy because of renal dysfunction. The risk of secondary malignancies in long-term survivors was not reported.

In our view, although high-dose chemotherapy with autologous hematopoietic stem cell rescue is a reasonable approach for children with high-risk localized disease who are treated according to the European induction regimen, these results cannot be extrapolated or incorporated into the standard North American treatment approach of intensively timed chemotherapy because the effect, outcome, and toxicity of such an approach are not addressed by the Euro-EWING trial results. Just as intensively timed chemotherapy is the currently favored strategy to intensify therapy in North America, the Euro-EWING results indicate that other strategies to intensify therapy (ie, high-dose chemotherapy with autologous stem cell rescue) can also improve outcomes in the context of the less intensive European induction regimen [52].

Local treatment — Local control for EWS can be achieved by surgery, RT, or both. The choice of RT or surgery usually represents a tradeoff between functional result and the risk of a secondary radiation-induced malignancy. As such, it should be decided through consideration of patient characteristics, the potential harm and benefit of the treatment options, and patient preference. In most cases of extremity EWS of bone, limb-salvage resections and reconstructions similar to those used for osteosarcomas have been used most frequently if it is anticipated that a wide margin can be achieved. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)

Patients who might lose function from surgical procedures because of tumor location (such as the periacetabular region of the pelvis) or extent may be offered RT as an alternative. However, advances in prosthetic reconstruction such as 3-D printed, custom prostheses even for the periacetabular area have become available to reconstruct surgical defects [53-55]. Data are limited for the long-term outcomes of these prosthetic reconstructive procedures. One observational study of 35 patients with pelvic EWS suggested that preoperative radiotherapy and surgical resection was associated with better histologic response and overall survival than either surgery alone or in combination with postoperative radiation [56]. The functional outcomes were better with hip transposition (ie, no implant used in the reconstruction) than if an implant were used, likely due to a much lower incidence of infection in those patients without an implant. However, this study is limited by the small number of patients.

Surgery is preferred for potentially resectable lesions and for those arising in dispensable bones (eg, fibula, rib, small lesions of the hands or feet) for the following reasons:

●It avoids the risk of secondary radiation-induced sarcomas. (See "Radiation therapy for Ewing sarcoma family of tumors".)

●An analysis of the degree of necrosis in the excised tumor can permit refinements in the estimate of prognosis. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management", section on 'Neoadjuvant chemotherapy'.)

●In the skeletally immature child, resection may be associated with less morbidity than RT, which can retard bone growth and cause deformity.

●Technical advances in the prostheses used for limb reconstruction of other bone sarcomas have led to improved long-term outcomes and can be applied to patients with Ewing sarcoma as well. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management", section on 'Reconstruction techniques'.)

Although there are no randomized trials comparing surgery with RT for local control, multiple retrospective series and a systematic review suggest superior local control and, in some cases, EFS for surgery compared with RT alone [19,57-60]. However, selection bias accounts for at least some of these results (ie, smaller, more favorably situated peripheral tumors or those that respond well to chemotherapy are more likely to be resected, while larger, axial lesions are radiated). The difference in local control between RT and surgery has been abrogated in some series when age and primary tumor site were controlled for [61-63]. Radiation dose and proper field planning are also important factors in local control. The role of RT in the local management of Ewing sarcoma is discussed separately. (See "Radiation therapy for Ewing sarcoma family of tumors".)

For primary tumors of the spine, complete surgical resection with negative margins is rarely feasible. Instead, definitive RT is usually the preferred mode of local control in these cases [64]. For the rare patients who are treated with surgery, the majority will require postoperative RT to achieve local control. (See 'Adjuvant radiation therapy' below.)

Patients with pelvic primaries have a high complication rate if treated surgically and poorer function if the acetabulum is resected [65], but some studies suggest superior survival with surgical resection. Local control of pelvic primaries must be individualized. There have been improvements in implants and spine fixation techniques that have potentially improved functional outcome following resection of selected primary tumors of the spine and pelvis. The experience in resecting other tumors of the spine and pelvis may be applied to EWS, but it requires careful selection of patients and a specialized center with experience in these reconstructions.

The surgical principles that apply to resection of the primary tumor and reconstruction are similar to those in patients with osteosarcoma and are discussed elsewhere (see "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management"). However, there are three main differences between EWS and osteosarcoma:

●EWS are radiosensitive, while osteosarcoma is much less so.

●EWS tend to occur in a younger population, where skeletal immaturity and concern for radiation-induced growth inhibition must be considered.

●EWS tend to arise in different areas of the long bones than osteosarcoma (the diaphysis compared with metaphysis (figure 1 and figure 2)) [66].

Adjuvant radiation therapy — RT is usually avoided in patients without residual disease (ie, resected with negative margins) to avoid exposing them to the risk of a radiation-induced malignancy. However, RT is an essential component of therapy for patients undergoing resection if the surgical margins are inadequate, although effective chemotherapy can also reduce the risk of local failure in such patients [9,15]. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Adjuvant radiation therapy'.)

TREATMENT FOR METASTATIC DISEASE — Patients with overt metastatic disease at presentation have a significantly worse outcome than do those with localized disease. However, aggressive multimodality therapy can relieve pain, prolong the progression-free interval, and cure some patients of their disease. In a review of 13 different series in which patients with metastatic EWS were predominantly treated with chemotherapy, five-year event-free survival (EFS) and overall survival rates averaged 25 (range 9 to 55) and 33 (range 14 to 61) percent, respectively [7,11,67-77]. The small numbers of patients in each series and the heterogeneity in location and extent of metastatic disease probably account for these wide variations in outcome. (See "Clinical presentation, staging, and prognostic factors of the Ewing sarcoma family of tumors", section on 'Disease extent'.)

In particular, the site of metastatic disease is an important variable. For patients with isolated lung and pleural metastases, EFS rates up to 40 percent are reported with multimodality therapy; for metastases involving bone or bone marrow, EFS rates fall to 10 to 20 percent and, for combined sites, to less than 15 percent [75]. Because it can be difficult to predict which patients with metastatic disease will be long-term relapse-free survivors [68], treatment should be administered with curative intent. (See 'General treatment principles' above.)

Several issues are pertinent to patients with metastatic EWS:

●What is the optimal initial chemotherapy regimen?

●Does dose intensification or interval compression provide benefit in the initial treatment phase?

●Is there a role for high-dose chemotherapy with stem cell rescue (termed autologous transplant) as consolidation after initial therapy?

●Is radiation therapy (RT) to sites of metastatic involvement of any benefit?

●How should control of the primary site be approached?

Standard chemotherapy — Patients with disseminated disease at diagnosis often respond well to the same type of systemic chemotherapy as is used for localized disease. Fewer randomized trials have been conducted in this population as it accounts for only 25 to 30 percent of patients with EWS. Therefore, there are fewer definitive conclusions that can be garnered from the published literature regarding a standard regimen. Whenever possible, patients with newly diagnosed metastatic EWS should be prioritized for enrollment in open clinical trials evaluating novel approaches. In the absence of an open clinical trial, regimens similar to those used for the treatment of patients with newly diagnosed localized disease are commonly used (see 'Chemotherapy' above):

●VDC plus standard-dose I/E – In contrast to the experience for patients with nonmetastatic disease, specific benefit for the addition of ifosfamide plus etoposide (I/E) to the vincristine, doxorubicin, and cyclophosphamide (VDC) backbone (VDC/IE) has not been shown for patients with metastatic disease at diagnosis [42,44,70,71,78,79]. Nevertheless, the combination of VDC/IE is commonly employed for this group of patients.

●Dose-intensified VDC/IE – In a nonrandomized trial, patients with newly diagnosed metastatic Ewing sarcoma were treated with VDC/IE with augmented alkylator doses [80]. Outcomes were similar to patients treated with non-intensified therapy. As a result, this approach has not been propagated further in North American trials for this population. (See 'Dose-intense chemotherapy' above.)

●Interval-compressed VDC/IE – There is little experience with interval-compressed chemotherapy in the context of metastatic EWS, although this regimen is frequently used in this context. (See 'Dose-intense chemotherapy' above.)

•Addition of ganitumab (IGF-1R antibody) – Based upon data demonstrating activity of the IFG-1R monoclonal antibody ganitumab in patients with relapsed disease [81], a randomized phase III Children's Oncology Group (COG) trial (AEWS1221) evaluated the addition of ganitumab to interval-compressed VDC/IE. However, preliminary results from the trial demonstrated no improvement in outcomes and increased toxicity with this approach [82].

Surgery and radiation therapy — Outcomes are best when chemotherapy is combined with optimal local therapy, including radiation and sometimes resection of sites of gross metastatic disease [76,83-85]. This multimodality approach has been most successful for patients with limited pulmonary metastases.

Pulmonary metastases — For patients with pulmonary metastases, we suggest a multimodality approach that includes chemotherapy and supplemental low-dose whole-lung irradiation, with surgical resection reserved for lung metastases that do not resolve with chemotherapy.

Patients with a limited number of lung metastases do not share the same dismal prognosis as patients with metastatic disease at other sites (ie, bone or bone marrow). Surgical resection may be undertaken in selected patients [86]. Although complete resection may be possible, chemotherapy is a necessary component of therapy, and five-year survival between 20 and 40 percent can be achieved [75,87,88].

Supplemental whole-lung radiation also appears to provide a benefit as a consolidation approach for patients with pulmonary metastatic disease, even after a complete response to chemotherapy. Most of the available data supporting radiation in this setting are from nonrandomized series, and patient selection factors confound interpretation of the data. Nevertheless, low-dose bilateral whole-lung irradiation (12 to 18 Gy in daily 1.5 to 2.0 Gy fractions) may provide added disease control in the lungs without significant pulmonary toxicity and is usually recommended after chemotherapy. This topic is discussed in detail elsewhere. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Pulmonary metastases'.)

Bone and soft tissue metastases — Patients with solitary or circumscribed bone or soft tissue lesions can be irradiated at those sites, usually to doses of 45 to 56 Gy, in addition to local control of the primary tumor. However, the likelihood of long-term survival is considerably lower than for patients with isolated pulmonary metastases. More attention has focused on the role of stereotactic body radiation therapy (SBRT) to oligometastatic sites of disease [89]. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Bone and soft tissue metastases'.)

Local control of the primary tumor — Local control can pose major issues for patients with metastatic disease. The complete response rate to initial chemotherapy can be increased with subsequent RT or selected excision for all sites of evident disease; however, long-term relapse-free survival is still only achieved in the minority of patients [68].

The presence of diffuse metastatic disease can make it difficult to justify a large resection, which would necessitate a lengthy period off systemic chemotherapy. For selected patients, resection of the primary tumor may be reconsidered if chemotherapy results in significant volume reduction, particularly if areas of small-volume metastatic disease are also amenable to surgical resection. On the other hand, RT, which is more often considered in patients with metastatic disease, will usually provide adequate local control with acceptable morbidity. If substantial amounts of bone marrow will need to be included in the radiation treatment volume, then radiating the primary tumor may occur first, followed by metastatic site radiation, delayed until the end of systemic therapy to avoid interfering with chemotherapy. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Management of the primary site'.)

Role for high-dose chemotherapy with stem cell support? — High-dose chemotherapy with hematopoietic stem cell support has no established role in the treatment of patients who present with metastatic disease, as this approach does not improve EFS or overall survival and is associated with significant toxicity [90,91]. Patients should be encouraged to participate in clinical trials evaluating other novel therapeutic approaches.

High-dose chemotherapy with hematopoietic stem cell support was evaluated in a randomized clinical trial (European Ewing Tumour Working Initiative of National Groups [Euro-EWING] 99 and EWING 2008) [90]. In this trial, 287 patients with isolated pulmonary (lung or pleural) metastatic disease received six cycles of vincristine, ifosfamide, doxorubicin, and etoposide (VIDE) and one cycle of vincristine, dactinomycin, and ifosfamide (VAI). Subsequently, patients were randomly assigned to receive either one course of busulfan plus melphalan high-dose chemotherapy followed by autologous stem cell rescue, or seven cycles of conventional chemotherapy with VAI followed by whole-lung irradiation. At a median follow-up of approximately eight years, there was no statistically significant difference in EFS between the two groups (eight-year EFS 53 versus 43 percent, hazard ratio [HR] 0.79, 95% CI 0.56-1.10), and overall survival results were comparable (eight-year overall survival 55 versus 54 percent, HR 1.00, 95% CI 0.70-1.44). Additionally, rates of infection as well as gastrointestinal and liver toxicities were higher in the high-dose chemotherapy group and included four deaths versus no deaths in the conventional chemotherapy group.

Previous observational data were mixed regarding the efficacy and toxicity of hematopoietic stem cell support [92-97], mostly due to variability in the definitions of high-risk patients, in metastatic disease sites, and in regimens. As examples, two prospective observational studies reported EFS rates as high as 50 percent with high-dose chemotherapy and hematopoietic stem cell support among patients with either isolated pulmonary metastases or multifocal disseminated disease [98,99]. In contrast, other studies evaluating this approach in patients with bone or marrow metastases showed less favorable results, with EFS ranging between 14 and 36 percent [100-105].

Based on the sum of data, we do not offer high-dose chemotherapy with hematopoietic stem cell support for those with metastatic EWS.

POSTTREATMENT SURVEILLANCE — There are no prospective data that address the appropriate schedule or selection of tests for surveillance for patients with EWS after initial treatment for localized disease. Consensus-based guidelines from the National Comprehensive Cancer Network (NCCN) [106] and from the Children's Oncology Group [107] recommend physical examination, a complete blood count, chest imaging, and surveillance imaging for primary site and distant recurrences every three months for two years, every six months for years 3 to 5, and annually thereafter (table 2) [106,108]. For local imaging of the primary site, we perform computed tomography (CT) scans for most patients. Magnetic resonance imaging (MRI) of the primary site is not usually necessary, and when obtained, it can offer little in the way of diagnostic information if a metal endoprosthesis is present. For surveillance imaging of potential distant metastases, we perform a chest radiograph, with subsequent evaluation using chest CT for abnormal imaging. Patients with symptoms or abnormal imaging or those who are planning surgical interventions may also be offered bone scan or fludeoxyglucose (FDG) positron emission tomography (PET)-CT, depending on whether these studies detected disease prior to therapy. A meta-analysis study showed that FDG-PET and PET-CT have a high accuracy in detecting distant metastases and postoperative recurrences in patients with EWS [109].

The appropriate duration of follow-up is unknown. Given the possibility of late relapse (although the vast majority of recurrences are observed within 10 years) and late development of treatment-related complications such as second neoplasms, some suggest that the patient be followed indefinitely [110]. There are no definite protocols for this, but the authors suggest that survivors have ongoing surveillance for late effects or late relapse. (See 'Evaluation' below.)

Physicians performing posttreatment surveillance must be cognizant of concerns for radiation exposure and the risk for secondary malignancies, particularly in younger individuals. (See 'Complications in long-term survivors' below and "Radiation-related risks of imaging".)

RECURRENT DISEASE — The majority of relapses occur within two years of initial diagnosis, but late relapse is not uncommon [111-113]. In a report from the Childhood Cancer Survivor Study (CCSS), the 20-year cumulative incidence of a late recurrence among five-year survivors of Ewing sarcoma was 13 percent [113]. For this reason, it is advisable that patients be followed for the potential of late relapse indefinitely. (See 'Posttreatment surveillance' above.)

In general, survival after an early relapse is poor, with few survivors among those who relapse within two years of therapy. In contrast, up to 15 to 20 percent of those who relapse later may survive long-term [114,115]. Other prognostic factors for death in patients with recurrent Ewing sarcoma include recurrence at combined local and distant sites, and an elevated lactate dehydrogenase (LDH) at initial diagnosis [114,115].

Evaluation — Symptoms at the primary site or elsewhere should raise concern and be appropriately investigated. Patients with a suspected recurrence should undergo evaluation both of the primary site and for the presence of metastatic disease before a treatment plan is formulated. The majority of patients with a local recurrence have either gross or microscopic metastatic disease.

The documentation of a local recurrence can be difficult. In patients with metallic endoprostheses, magnetic resonance imaging (MRI) and computed tomography (CT) evaluation can be distorted by metal artifact, although this is improved with contemporary MRI and CT imaging technology. In patients who are treated with certain extendable "growing" prostheses, MRI may be contraindicated. The interpretation of irradiated areas on imaging studies can be challenging because of the changes in bone caused by the radiation. Soft tissue masses may represent residual fibrosis rather than recurrent tumor. The evaluation of intraosseous sites is even more difficult since the response to prior therapy and the variability in reossification complicate the interpretation of radiographic studies. Progressive cortical destruction or increasing radiolucent areas suggest local recurrence, as do bone scans that demonstrate increased radiotracer uptake. Positron emission tomography (PET) scans may be useful to assess the likelihood of recurrence at a site that is suspicious on cross-sectional radiographic imaging. Ultrasound can also be useful to assess for soft tissue masses around an endoprosthesis. The decision to biopsy a site of suspected recurrent local or metastatic disease depends upon patient history and the level of evidence from imaging studies. Open bone biopsies can be associated with local morbidity (ie, wound complications and bone fracture), and needle biopsies may suffice.

Treatment — Although the prognosis for patients with recurrent disease is poor, some patients can be successfully salvaged, particularly patients with late relapses [116,117]. Sites of recurrence, prior treatment, and relapse-free interval affect remaining treatment choices.

Most patients with recurrent disease will receive systemic therapy prior to attempts at additional local control measures. For patients with an initial recurrence, a camptothecin-based regimen (irinotecan/temozolomide or topotecan/cyclophosphamide) is most commonly used [118-120]. Patients relapsing after a lengthy disease-free interval off chemotherapy may respond again to the same agents used as part of initial therapy. The combination of gemcitabine and docetaxel has also been used in this context, with variable results.

In a randomized trial (rEECur) of 336 patients with relapsed or refractory EWS, four different chemotherapy regimens are being directly compared: irinotecan plus temozolomide; topotecan plus cyclophosphamide; high-dose ifosfamide; and gemcitabine plus docetaxel [121]. The first two interim analyses demonstrated that the irinotecan plus temozolomide and gemcitabine plus docetaxel arms are less effective than the remaining arms. As a result, these two arms have been dropped from further study and comparison is ongoing for efficacy of the topotecan plus cyclophosphamide and high-dose ifosfamide arms.

Local management of a local recurrence usually includes surgery (and possibly an amputation if the local recurrence involves an irradiated extremity), radiation therapy (RT), or both. RT to bone lesions usually provides pain relief, while surgery can eradicate disease in some cases with limited isolated lung metastases [87]. As an example, one study of 26 patients with local recurrence showed that the survival at five years post local recurrence was 28 percent. Better survival outcomes were seen in those who did not have metastases at diagnosis of the recurrence, had a surgical treatment for the recurrence, and had complete eradication of all disease [122].

Investigational agents — Patients with recurrent or advanced EWS should be encouraged to participate in clinical trials, where available. Future therapies will likely emerge as the fundamental biology of the fusion oncoproteins driving this disease is better understood [17,123]. (See "Epidemiology, pathology, and molecular genetics of the Ewing sarcoma family of tumors" and "Clinical presentation, staging, and prognostic factors of the Ewing sarcoma family of tumors", section on 'Prognostic factors'.)

Cabozantinib, an inhibitor of the vascular endothelial growth factor receptor (VEGFR) and MET signaling pathways, has demonstrated activity in patients with advanced Ewing sarcoma, with an objective response rate of 26 percent and median progression-free survival of five months in one phase II trial [124]. However, its use remains investigational, and further studies are required to confirm its efficacy.

Examples of other drugs under investigation include agents targeting the insulin-like growth factor 1 (IGF-1) receptor [81,125-128], poly(ADP-ribose) polymerase (PARP) inhibitors [129-131], lysine specific demethylase 1 (LSD1) [132], and RNA helicase [133].

COMPLICATIONS IN LONG-TERM SURVIVORS — Although the survival of patients with EWS has improved steadily since the 1970s, long-term survivors have considerable burden of the late effects of their therapy [16,61,110,134-136]. These include subsequent primary cancers, pathologic fractures, other radiation-associated complications (wound complications, pulmonary fibrosis, neuropathy, limb leg discrepancy, femoral head necrosis), and chemotherapy-related complications (subsequent primary cancers, reduced fertility, renal insufficiency, and cardiomyopathy) [134,137].

Secondary myelodysplasia (MDS) and leukemia are particular concerns for this population [102,105,138-140]. This was illustrated in a report from the Children's Oncology Group (COG) of 578 children with EWS who were treated with three different regimens over a six-year period [138]. Overall, 11 children developed secondary MDS/acute myeloid leukemia (AML), and the cumulative risk was significantly higher among children treated with a regimen incorporating higher doses of doxorubicin, cyclophosphamide, and ifosfamide as compared with those receiving standard-dose vincristine, doxorubicin, cyclophosphamide, and dactinomycin (VDCA) with or without ifosfamide plus etoposide (11, 0.9, and 0.4 percent at five years, respectively).

The health status of long-term (≥5 years) survivors was addressed in a cohort study of 568 individuals who were diagnosed with EWS before age 21 from 1970 to 1986, including a subset of 403 patients who were participating in the Childhood Cancer Survivor Study (CCSS) [136]. Cumulative mortality among all survivors was 25 percent at 25 years after diagnosis, and the cumulative incidence of secondary malignancy was 9 percent. Disease progression/recurrence accounted for 60 percent of all deaths, while other causes included secondary neoplasms, cardiac disease, other medical causes, and pulmonary disease. The cumulative mortality attributed to subsequent malignant neoplasms and cardiopulmonary disease potentially attributed to treatment was 8.3 percent at 25 years. In addition, compared with their siblings, survivors had significantly higher rates of severe, disabling, or chronic health conditions, significantly lower fertility rates, and higher rates of self-reported moderate to extreme adverse health status.

It is anticipated that the adoption of tailored radiation ports in the last 15 years (as opposed to whole-bone ports used in the period 1960 to 1980), the use of lower and risk-adopted radiation therapy (RT) doses, an appreciation of the rise in radiation-induced subsequent primary cancer risk with doses above 60 Gy, and the increased use of surgical resection in the population with nonmetastatic disease will all reduce the risk of late effects. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Late effects'.)

Nevertheless, the relatively high complication rates seen with many of these earlier treatment approaches, the delayed nature of many of the complications, and the possibility that trends in chemotherapy intensification may alter the pattern of secondary malignancies [135] underscore the need for long-term follow-up. Long-term follow-up guidelines after treatment of childhood malignancy are available from COG.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Soft tissue sarcoma".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Ewing sarcoma (The Basics)" and "Patient education: Bone cancer (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Patients with Ewing sarcoma (EWS) require referral to centers that have multidisciplinary teams of sarcoma specialists. With rare exception, systemic combination chemotherapy and definitive local therapy are required in all patients, and care should be coordinated among the medical/pediatric oncologist, surgeon, and radiation therapist. (See 'General treatment principles' above.)

Treatment of adults with EWS should be guided by the same general principles as are used for younger individuals. (See 'Adult patients' above.)

●In most cases, treatment will begin with chemotherapy. For most children and adults with localized EWS who are treated in North America, we recommend interval-compressed therapy with alternating cycles of vincristine/doxorubicin/cyclophosphamide (VDC) and ifosfamide/etoposide (VDC/IE) given every two weeks with hematopoietic growth factor support (table 1), rather than the same regimen given every three weeks without growth factor support (Grade 1B). Outside of North America, other strategies to intensify therapy (ie, consolidative high-dose chemotherapy with autologous stem cell rescue) can also improve outcomes in the context of the less intensive European induction chemotherapy regimen. (See 'Dose-intense chemotherapy' above.)

●Local control for EWS can be achieved by surgery, radiation therapy (RT), or both. The choice of RT or surgery usually represents a tradeoff between functional result and the risk of a secondary radiation-induced malignancy. As such, it should be decided through consideration of patient characteristics, potential harm and benefit of the treatment options, and patient preference. For most patients with nonmetastatic disease, surgical resection is preferred if it is anticipated that a complete resection with negative margins can be achieved and a functional reconstruction is possible. Patients who lack a function-preserving surgical option because of tumor location or extent may be offered RT as an alternative to resection. Surgery is also preferred for lesions arising in dispensable bones (eg, fibula, rib, small lesions of the hands or feet). A combination of radiation and surgery is reserved for cases in which negative margins cannot be achieved while still preserving function. (See 'Local treatment' above and "Radiation therapy for Ewing sarcoma family of tumors", section on 'Adjuvant radiation therapy'.)

Following local treatment, chemotherapy is usually continued (the same alternating VDC/IE regimen), typically for several months (table 1).

●Patients with clinically detectable metastatic disease at initial presentation also require multimodality therapy. Patients with advanced disease typically should be approached with potentially curative treatment. Up to 40 percent of patients with limited pulmonary metastatic disease who undergo intensive chemotherapy and pulmonary resection with or without RT may be long-term survivors. The prognosis for other subsets of patients with advanced disease is less favorable. (See 'Treatment for metastatic disease' above.)

For patients with pulmonary metastases, we suggest a multimodality approach that includes chemotherapy and supplemental low-dose whole-lung irradiation (Grade 2C), with surgical resection for lung metastases that do not resolve with chemotherapy. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Pulmonary metastases'.)

There is no conclusive evidence that high-dose chemotherapy with hematopoietic stem cell infusion is beneficial for patients with metastatic EWS, and it is not our standard approach. (See 'Role for high-dose chemotherapy with stem cell support?' above.)

●The majority of relapses occur within two years of initial diagnosis, but late relapse is not uncommon. For all EWS of bone or soft tissue, there are no prospective data that address the appropriate schedule or selection of tests for surveillance after initial treatment for localized disease. Consensus-based guidelines from the National Comprehensive Cancer Network (NCCN) and from the Children's Oncology Group (COG) (table 2) recommend physical examination, a complete blood count, chest imaging, and local imaging of the primary site every three months for two years, every six months for years 3 to 5, and annually thereafter. Long-term follow-up (lifelong) is needed following therapy because disease relapse, treatment-related complications, and second malignancies all occur beyond five years after treatment is initiated. (See 'Posttreatment surveillance' above.)

●Although the prognosis for patients with recurrent disease is poor, some patients can be successfully salvaged. The sites of recurrence, prior treatment, and relapse-free interval affect the remaining treatment choices. Most patients with recurrent disease will receive systemic therapy prior to attempts at additional local control measures. (See 'Recurrent disease' above.)

●The survival of patients with EWS has steadily improved; however, long-term survivors have considerable burden of the late effects of their therapy. These include subsequent primary cancers, pathologic fractures, other radiation-associated complications (wound complications, pulmonary fibrosis, neuropathy, limb leg discrepancy, femoral head necrosis), and chemotherapy-related complications (subsequent primary cancers, reduced fertility, renal insufficiency, and cardiomyopathy). Long-term follow-up guidelines after treatment of childhood malignancy are available from the COG. (See "Radiation therapy for Ewing sarcoma family of tumors", section on 'Late effects'.)

ACKNOWLEDGMENT — The editorial staff at UpToDate would like to acknowledge David C. Harmon, MD, who contributed to an earlier version of this topic review.