Ewing Sarcoma

Etiology

There is no well-established association between Ewing sarcoma and environmental risk factors, drug exposure, radiation history, or cancer history in the family. Studies have been limited to small retrospective, case-control studies.

Epidemiology

Ewing sarcoma is the second most common primary bone malignancy in adolescents and young adults with a median age of 15 years and accounts for less than 5% of all soft tissue sarcomas. There are more than 200 cases per year in the United States. The incidence of Ewing sarcoma in the United States was 2.93 per million between 1973 and 2004.[4]

The peak incidence is between 10 and 15 years, with around 30% of the cases arising in children under 10 and another 30% in adults over 20. There is a male predominance with a male-to-female ratio of 3 to 1. Whites are much more frequently affected than Blacks, Asians, Hispanics, or Africans. This significant racial discordance has yet to be explored. The actual incidence of Ewing sarcoma in older populations is unknown.

Pathophysiology

The cell of origin for Ewing sarcoma is not fully understood. Nonrandom gene rearrangements involving the EWS and the ETS (E26 transformation-specific or E-twenty-six) gene family distinguish Ewing sarcoma.[5] The most frequent gene rearrangement is t(11;22)(q24;q12). A hybrid gene EWS–FLI1 generated by the fusion of the EWS gene on 22q12 with the FLI1 gene on 11q24 occurs in greater than 80% of reported cases. The resulting EWS-FLI1 fusion protein acts as an aberrant transcription factor, which may play a role in the pathogenesis of Ewing sarcoma.[5] The translocation's underlying etiology has not been established.

The second most common EWS–ETS gene family rearrangement is t(21;22)(q22;q12) translocation, resulting in the fusion of EWS with the ERG gene on 21q22 (observed in about 15% of cases). In the literature, other detected gene rearrangements are t(7;22)(p22;q12), t(17;22)(q21;q12) and (t(2;22)(q33;q12) where EWS is fused with ETS gene family ETV1, E1AF and FEV, respectively.[6] More chromosomal translocations and complex gene rearrangements have also been reported in the literature. However, whether they are associated with more aggressive tumor features is unclear.

Histopathology

Ewing sarcoma is composed of small round cells with an increased nuclear-cytoplasmic ratio that represents a family of small round blue cell tumors of childhood (eg, retinoblastoma, neuroblastoma, rhabdomyosarcoma, and nephroblastoma). Ewing cells have scant eosinophilic cytoplasm containing abundant glycogen often detected by staining with periodic acid-Schiff.[2]

More than 80% of the cases exhibit high CD99 expression. This highly sensitive immunohistochemical biomarker likely plays a key role in facilitating the continued migration of leukocytes to endothelium; however, this biomarker lacks specificity, as it may also be detected in other sarcomas and lymphomas.[2] In addition to MIC2 gene product CD99, Ewing cells often express CD45, synaptophysin, chromogranin, vimentin, keratin, desmin, neuron-specific enolase, and S-100. However, this immunohistochemistry panel has also been limited by a need for more specificity.[2] Molecular genetic studies using fluorescence in situ hybridization and reverse transcription-polymerase chain reaction are required to make the definitive distinction.

History and Physical

Patients with Ewing sarcoma often present with local symptoms such as pain, stiffness, or swelling for a few weeks or months. More than 50% of patients with Ewing sarcoma have intermittent pain that worsens at night. Ewing sarcoma is commonly found in the diaphysis of long bones but can occur in different locations with varying presentations. Bone or metastatic lesions within the long bone can present as pathological fractures. The pelvic location of Ewing sarcoma can present as back pain. The presence of systemic symptoms, including fever and weight loss, often indicates metastatic disease. Around 20% of patients present with metastatic disease at the time of diagnosis, and among these cases, more than 20% have lung or pleura involvement.[7]

A comprehensive physical examination is critical, as is a neurologic examination in patients with CNS involvement. Patients can present with asymmetric breath sounds, pleural signs, or rales in lung and pleura metastasis. Petechia or purpura from thrombocytopenia can be observed in patients with bone marrow metastases. A neurologic examination is also critical in patients with CNS involvement. Initial workups include an x-ray of the affected area that may show the "onion skin" appearance of periosteal reaction.[8] A bone scan, magnetic resonance imaging, and computed tomography scan are required for initial staging.

Evaluation

Imaging tests should evaluate the primary site and potential metastatic sites. Plain radiographs of the affected area may show destructive confluent '' moth-eaten" lesions, "Codman triangle" of the elevated periosteum, or multilayered "onion-skin" periosteal reaction. Imaging of primary sites includes MRI with or without CT, with contrast is of prime importance as this allows for determination of the extent of disease, operability, degree of edema, and adjacent organ involvement.[9] Other imaging methods, including CT thorax, positron emission tomography (PET)/CT, bone scan, and MRI of the spine and pelvis, can also be used to detect possible lymph node involvement or metastatic sites.[10] 

Patients who are symptomatic should be referred to the orthopedic surgeon if a biopsy is needed. An open biopsy provides adequate sampling, even though a CT-guided core needle biopsy can establish the diagnosis. The diagnostic workup should include molecular cytogenetic analysis of biopsy specimens to evaluate the t(11;22) translocation.[6] Bone marrow aspiration with smear and bone marrow biopsy should be considered. The first assessment should consist of serum lactate dehydrogenase, which carries prognostic significance. Patients should also be offered fertility counseling before starting the treatment.[11]

Treatment / Management

The standard of care for patients with or without metastasis includes interprofessional treatment with chemotherapy and local therapy, including surgery and radiotherapy. Broadly, systemic therapy is the cornerstone of treatment for all Ewing sarcoma patients. In the United States, this consists of VDC (vincristine/doxorubicin/cyclophosphamide) with alternating IE (ifosfamide/etoposide). After induction chemotherapy, clinicians typically recommend local therapy using radiation, surgery, or a combination of both. Nonmetastatic disease has a 5-year survival rate of 75% to 80%, while metastatic disease is around 30%.   

Differential Diagnosis

Combining immunohistochemical and morphological findings with relevant clinical history can help narrow down the differential diagnosis of Ewing sarcoma that includes other small round cell tumors such as neuroblastoma, rhabdomyosarcoma, lymphoma, neuroectodermal tumors, desmoplastic small round cell tumor, and synovial sarcoma. Also, consider osteomyelitis, osteogenic sarcoma, and eosinophilic granuloma in the differential diagnosis of Ewing sarcoma.

Surgical Oncology

Surgical resection of Ewing sarcoma is considered a cornerstone of local therapy; resection is diagnostic and used to evaluate tumor response. Moreover, this treatment modality is believed to entail fewer long-term morbidities than radiotherapy (RT), particularly concerning growth restriction and the development of secondary malignancies. The use of amputation in this setting has declined significantly, and function-preserving surgery with custom endoprosthesis has increased—especially with the advent of 3D printing systems.[12][13] 

Surgery is the preferred method of local control, especially in tumors arising from expendable bones, including the ribs, fibula, iliac wings, distal four-fifths of the clavicle, and scapula. No prospective randomized trials comparing radiotherapy to surgery have been performed; however, a meta-analysis of several prospective trials suggests that, compared to surgery, only local control may be superior with postoperative radiotherapy and surgery in situations following intralesional resections and when tumors have wide resections or poor histologic response.[14] Data from patients enrolled in INT-0091, INT-0154, or AEWS0031 studies were analyzed by using the database of Clusters of Orthologous Groups of proteins, which also showed surgery plus RT was associated with a lower risk of local failure than definitive RT.[15][16]

Biopsy

An open incisional biopsy or core needle biopsy is recommended, depending on the treatment protocol. While a core needle biopsy can be adequate for diagnosis, the growing reliance on biologic studies demands additional tissue, which may not be adequately obtained through multiple needle biopsies. However, consensus guidelines continue to allow for either approach.[17] Biopsy tracts are routinely resected during surgery, as is common with other sarcoma types.[17] Areas of suspected metastasis, including lymph nodes or bone lesions, should also undergo biopsy.

Limb Sparing Surgery

Preoperative imaging is critical for determining the resectability of these tumors. MRI can aid in determining tumor extent as well as the amount of peri-tumoral edema. PET/CT can identify involved lymph nodes and metastatic disease. Anatomic considerations are critical to ensuring the complete removal of disease and minimizing postoperative morbidity.

Sacral tumors that cross the midline are generally unresectable, as are tumors that require the removal of a visceral organ.[17] Debulking surgeries are not advised in patients with inoperable disease.[17] Tumors of the spine may be resected, but only in carefully selected patients who respond well to chemotherapy. Patients with cord compression and neurological compromise need urgent surgical decompression treatment.[17] However, radiotherapy or chemotherapy can also be considered. Tumors of the chest wall, particularly the ribs, require resection of the involved chest wall, pleura, and typically a rib above and below the tumor. Ewing sarcoma of the scapular body without chest wall involvement can be done with little or no functional compromise. The sternum and lateral four-fifths of the clavicle can be easily resected without reconstruction. The proximal humerus and diaphyseal region may be resected but will likely require reconstruction with allografts or metallic implants. The radius and ulna may undergo an intercalary resection (removal of the diaphyseal portion with preservation of the native joint), but allografting or autografting may be required. Tumors of the skull base and mandible are complex and require input from multiple subspecialists.   

Once a patient qualifies as a surgical candidate, recommending wide excision of the primary tumor and peritumoral edema is advised. An adequate surgical margin entails a rim of normal tissue surrounding the tumor.

Amputation 

Amputations are infrequently performed but might be considered in cases where achieving a negative margin is not feasible with less radical approaches, in limbs with poor baseline function, or instances of locally recurrent disease.[17]

Metastatic Disease

In the setting of diffuse metastatic disease, surgical resection of the primary tumor may be difficult to justify. However, this may be reasonable for palliation or in cases where the metastatic disease burden is limited. While prospective data is lacking, a SEER database review of 643 patients who presented with metastatic Ewing sarcoma demonstrated improved 5-year overall survival and cancer-specific survival in patients who underwent local resection.[18]

While the metastatic disease has a poor prognosis, patients with isolated pulmonary metastasis may be amenable to resection. Metastatectomy in selected patients had a 3-year overall survival of 60% and event-free survival of 45%.[19] Candidates should be selected on a case-by-case basis.

Recurrent Disease

Approximately 70% of relapses occur within the first 2 years of diagnosis.[3] Although the most common sites of relapse are distant spread, 10% to 15% of patients may have an isolated local recurrence or a single metastatic site.[3] These patients may be considered for additional local therapy. A retrospective review of earlier Ewing sarcoma trials revealed that patients with local relapse who were able to undergo repeat resection had a 5-year probability of survival of 30% versus 9% in those who could not attain resection.[20] However, the small number of patients and selection biases must be acknowledged when considering treatment options for relapsed patients.

Complications 

Approximately 25% of patients experience complications that can be attributed to surgery directly or because of multimodality treatment.[21] These include delayed wound healing, fibrosis, thrombosis, abscess formation, or hardware loosening/failure.[21][22] The risk of postoperative complications appears to be related to the tumor site (eg, pelvis vs extremity).

Radiation Oncology

Radiotherapy is used for patients with Ewing sarcoma who cannot obtain negative margins or have inoperable or metastatic disease. While there have been no direct prospective trials comparing radiotherapy to surgical resection, retrospective data suggest local failure rates are better with surgery (4% vs 15%), and local failure appeared higher in those >18 years with pelvic tumors.[23] In patients requiring both surgery and radiotherapy, the local failure rate was 6.6%. No survival difference has been found between local therapy with surgery alone, radiotherapy alone, or in combination.[15][24] Radiotherapy has also been used in the metastatic setting for disease control and may improve event-free survival.[25] 

Preoperative Radiotherapy

The use of preoperative radiotherapy is typically encouraged in adults with sarcomas. This therapy provides smaller margins, a definitive target, and lower radiation doses than postoperative radiotherapy.[26] Despite these advantages, radiotherapy has yet to be prospectively investigated; therefore, it is not broadly recommended for Ewing sarcoma in pediatric patients as its use appears indent. For pelvic Ewing sarcoma, retrospective analysis of preoperative radiotherapy administered on a non-selective basis appeared to confer higher rates of 5 local recurrence-free survival (88% vs 66.5%), adequate surgical margins (81.5% vs 59.1%), and higher responses to chemotherapy (96.3% vs 63.6%) compared to a more conservative use.[27] However, there was no difference in metastasis-free survival or overall survival.[27] 

There are concerns regarding the widespread adoption of preoperative radiotherapy, including potential overtreatment, obscuring pathological assessment of chemotherapy response, as well as the long-term toxicity of this treatment.[27] Consensus guidelines indicate preoperative radiotherapy may be appropriate for patients with a poor response to chemotherapy, anticipated close/positive surgical margins, tumors close to critical structures, and large tumor volume.[17] Preoperative radiotherapy is allowed on current AEWS protocols.

Definitive Radiotherapy

Definitive radiotherapy is recommended for Ewing sarcoma patients with unresectable disease or where resection would be especially morbid. While radiotherapy has never been prospectively compared to surgical resection, this remains an option, especially for patients with bulky pelvic tumors. Ewing sarcoma is considered more radiosensitive than other sarcoma subtypes. Several retrospective reviews have demonstrated approximately 70% to 80% local control rates.[28][29][30] There have been suggestions that smaller lesions (<8 cm) treated with radiotherapy alone may have exceptionally high rates of local control (90% vs 52%).[31]

Adjuvant Radiotherapy 

Adjuvant radiotherapy is typically indicated in patients with microscopic or gross residual disease following surgical resection, intraoperative tumor spill, or poor responders to chemotherapy. Adjuvant radiotherapy is not recommended in patients who have undergone a complete resection. As with preoperative radiotherapy, evidence of benefit is not firmly established and the data are conflicting. In retrospective reviews, patients treated with adjuvant radiotherapy after inadequate resection had similar rates of relapse compared to inadequate resection alone (12% vs 14%).[32] However, the EURO EWING-99 study revealed that patients recommended for adjuvant radiotherapy but did not receive it had an 8-year local recurrence rate of 30%, contrasting with the 11% recurrence rate among those who followed the recommended treatment guidelines.[25] 

Intraoperative Radiotherapy 

Intraoperative radiotherapy is allowed on current protocols and delivered as a single fraction. The advantages of this treatment modality are shorter treatment times and limited treatment volume. Intraoperative radiotherapy can be given as a boost dose to reduce the number of EBRT treatments needed. This therapy is indicated for treating residual disease or recurrent disease in the operative tumor bed. Five-year local control rates were 70% to 77%.[33] The dose-limiting toxicities are peripheral neuropathy and soft tissue necrosis. 

Proton Therapy

The Bragg peak observed with proton therapy has the theoretical advantage of less normal tissue radiated, nonexistent exit dose, and less overall toxicity than photons. Proton therapy may also allow for dose escalation when needed.[34] The difficulty with evaluating proton therapy compared to photon is the lack of prospective randomized data. The evidence remains retrospective and subject to various biases. Pelvic, base of skull, chest wall, and axial skeleton tumors close to the spinal cord theoretically would benefit most from proton therapy. In women with pelvic sarcomas, ovarian doses were substantially lower with proton therapy compared to photon treatment.[35] 

The risk of secondary malignancy would also be reduced because of the reduced volumes of normal tissue irradiated compared with photons. A review of patients treated at the Harvard Cyclotron Laboratory from 1973 to 2001 was compared with patients treated with photons from the SEER database. After adjusting for sex, age at treatment, primary site, and year of diagnosis, the development of secondary malignancy was not higher with proton than with photon treatment (5.2% vs 7.5%).[36] A retrospective review of 30 patients with Ewing sarcoma treated with proton therapy found no late toxicities in the bowel, bladder, heart, and lung for the patients treated in the pelvic and thoracolumbar spine—which was attributed to the rapid dose falloff of protons.[37] 

Regarding oncologic outcomes, the 3-year local control with proton therapy was 92% with radiotherapy alone and 100% when combined with surgery.[34] These outcomes were comparable to other series using photon-based treatments.[34] Patients with non-metastatic Ewing sarcoma of the skull base treated with proton therapy had a 4-year local control rate of 96%.[38] However, this was not without late toxicities, such as bilateral hearing loss and vasculopathy.[38] While using proton therapy in children has theoretical advantages compared to photons, stereotactic body radiotherapy or brachytherapy may also offer highly conformal treatments and spare normal tissues and protons in many cases.

Metastatic Disease

In retrospective analysis, local treatment of disseminated disease with either surgery, radiation, or combination was shown to improve event-free survival (EFS). Radiotherapy was found to have a 3-year EFS of 23%, while patients without local treatment had an EFS of 13% over the same period.[25] One area where this is particularly true is in lung metastasis treatment. Whole lung irradiation is recommended in patients with any number of lung metastases regardless of response to chemotherapy. Attempts to replace whole lung irradiation with high-dose therapy have failed.[39]

Stereotactic body radiotherapy (SBRT) in patients with limited metastasis has been investigated. SBRT in the setting of pulmonary metastasis produced a partial response in 88% of patients without any grade 3 or worse toxicity.[40] Treatment of extrapulmonary metastasis with SBRT has also been investigated. SBRT to bony metastasis is well tolerated; however, significant toxicity can result based on location, radiation status, and the administration of chemotherapy.[41][42] Nonrandomized prospective trials have been attempted, but low enrollment makes it difficult to conclude (NCT02581384). SBRT is currently being investigated on the AEWS1221 protocol. 

Acute Toxicity

Acute toxicity of radiotherapy is highly dependent on the irradiated area and may consist of dermatitis, cystitis, mucositis, nausea, vomiting, or diarrhea. While these side effects can be troublesome for patients, they are typically temporary and resolve in the days to weeks following radiation. 

Late Toxicity

The late toxicities, which can occur months to years after treatment, are usually the most problematic for long-term survivors compared to the acute side effects. Like acute side effects, they largely depend on the irradiated location. 

Secondary Malignancies

Either chemotherapy or radiotherapy may cause secondary malignancies. Radiotherapy can induce osteosarcomas, although many other solid tumors have been reported. A recent systemic review of patients with Ewing sarcoma reported the cumulative incidence of secondary malignancies at 20.5% at 30 years, with 63% being solid tumors.[43] Retrospective analysis suggests that the cumulative risk of second malignancies is dose-dependent, with 0 cases at doses <48 Gy and 130 reported cases at doses exceeding 60 Gy.[44]

Pathological Fracture

Weight-bearing bones are at increased risk for pathological fractures. As a result, the femur is the most common site. The rate of fracture ranges from 20% to 33%.[45][46] If possible, half of the bone volume should not receive more than 50 Gy, and the 50 Gy isodose line should not encompass the entire circumference of the bone. 

Edema

Edema can develop in patients with Ewing sarcoma of the limbs where the entire circumference is irradiated; this leads to fibrotic changes causing lymphatic obstruction, further leading to edema in the extremity distally. Once developed, edema is irreversible. A strip of skin is typically spared during radiation planning, thus allowing for lymphatic drainage to reduce edema risk. This can be seen in approximately 20% of patients.[46]

Fibrosis/Loss of Mobility

Fibrosis can occur in up to 90% of cases and occurs at higher rates in postoperative radiotherapy compared to preoperative treatment.[46][47] Limb mobility impairment was seen in approximately 20% to 40% of patients.[46][47] Means of minimizing this toxicity are meticulous radiation planning and strict dose constraints. 

Wound Healing 

Patients undergoing preoperative radiotherapy have been shown to have higher rates of delayed wound healing post-op compared to those receiving adjuvant therapy (35% vs 17%).[26]

Growth Deficiency

The growth plates are moderately sensitive to radiation. Doses of 10 Gy to the epiphyseal plate can result in stunted growth, and a whole of >20 Gy can lead to complete arrest—which can lead to significant length discrepancies between the irradiated and unirradiated limbs. Younger patients (<15 years old) tend to have limb shortening, while those over 15 do not appear to be at risk, perhaps due to growth plate closure.[46] Limb atrophy may be found in up to 80% of patients.[46]

Timing

The timing of chemotherapy and radiotherapy depends on the clinical scenario. Typically, chemotherapy and radiotherapy are delivered concurrently. However, anthracyclines are typically held during radiation treatment. In emergencies such as cord compression, vision loss, or other acutely threatened organ function, the entire radiotherapy course may be delivered immediately on day 1. Although clinically, this scenario is not common. Usually, local therapy (surgery, radiotherapy, or both) is typically given in the adjuvant setting and delivered concurrently with consolidation chemotherapy at weeks 12 to 14. If pre-operative radiotherapy is planned, patients are treated at week 1 of consolidation therapy. If an additional boost is planned after surgery, it should commence within 2 weeks or as soon as the patient has recovered. Delays in local therapy exceeding 16 weeks were associated with worse 10-year overall survival outcomes (70.3% vs 57.1%).[48]

Dosing

The target dosing for Ewing sarcoma shown in Table 1 below was derived and modified from the AEWS1031 protocol and NCCN guidelines. If given preoperatively, the dosing is lower, but an additional boost may be required if the margins are inadequate. Postoperative doses are typically higher and require additional boost for inadequate margins. The doses are delivered in 1.8 to 2.0 Gy per fraction except for whole lung/abdominal radiation, which is delivered in 1.5 Gy per fraction.

Table 1. Target Dosing for Ewing Sarcoma

Dose Constraints

The dose constraints shown in Table 2 below are taken from the AEWS1031 protocol. In the pediatric population, efforts should be made to keep the dose to organs at risk as low as reasonably achievable to reduce the risk of late toxicity.  

Table 2. Organ Radiotherapy Dose Constraints, AEWS1031 Protocol

Target Delineation 

The initial gross tumor volume (GTV1) is outlined based on radiographic and palpable disease before any surgical resection or chemotherapy; this would include any clinically involved lymph nodes. Multiple imaging modalities may be used to aid in outlining gross disease, including MRI, PET, and CT scans that may be fused to the simulation CT. The initial clinical target volume (CTV1) will encompass the initial GTV1 and cover occult disease around the gross disease by adding a 1 to 2-cm margin and the draining lymph node chain surrounding any involved nodes. The CTV should be excluded, and natural barriers to spread should be respected. The initial PTV1 consists of the CTV1 with a 0.5 to 1.0 cm margin, depending on the availability and sophistication of the institution’s image guidance. Depending on the tumor location, motion management and creating an internal target volume may be needed. The boost volumes may be required. The gross target volume (GTV2) will encompass the pre-chemotherapy disease in bone but post-response to chemotherapy in soft tissue in unresected patients. In partially resected patients with macroscopic disease, the GTV2 includes gross residual tumor, tumor bed, and pretreatment disease in the bone. For microscopic residual disease, the GTV2 includes disease from post-induction chemotherapy imaging. The clinical target volume for the boost is a 0.5 to 1.5 cm margin around the GTV2. The PTV2 is constructed around the CTV2 with a 0.5 to 1.0 cm margin. Depending on the tumor location, modifications to these volumes may need to be made. 

Medical Oncology

Localized Disease

Historically, Intergroup Ewing Sarcoma Study trials (IESS-I and IESS-II) showed better results with RT plus adjuvant chemotherapy with VACA (vincristine, dactinomycin, cyclophosphamide, and doxorubicin) as compared to VAC (vincristine, dactinomycin, and cyclophosphamide). Due to the dose limitation of doxorubicin in dactinomycin regimens, trials after that demonstrated no significant impact on clinical outcomes with the omission of dactinomycin. Several studies evaluated the addition of ifosfamide and etoposide to standard chemotherapy. The Pediatric Oncology Group-Children’s Cancer Group study INT0091 demonstrated that the VACD-IE group had significantly better survival rates than the VACD group. Also, the VACD-IE group was associated with a lower incidence of local failure.[49]

In the European Intergroup Cooperative Ewing Sarcoma Study (EICESS-92), VACA (vincristine, dactinomycin, cyclophosphamide, and doxorubicin) and VAIA (vincristine, dactinomycin, ifosfamide, and doxorubicin) were compared in the standard risk patients, and the effect of cyclophosphamide was found similar to ifosfamide; however, cyclophosphamide was associated with increased toxicity.[50] The 3-year event-free survival rates were 73% and 74% for VACA and VAIA, respectively.

Euro-EWING99-R1 trial (noninferiority trial based on EICESS-92 protocol) evaluated whether cyclophosphamide could replace ifosfamide in consolidation therapy, including vincristine and dactinomycin in patients with standard-risk and suggested that VAC (vincristine, dactinomycin, and cyclophosphamide) was statistically not inferior to VAI (vincristine, dactinomycin, and ifosfamide); however, VAI was associated with slightly higher 3-year event-free survival.[51]

In a phase III trial (AEWS0031) from the Children’s Oncology Group (COG), patients on the standard arm received VDC alternating with IE every 3 weeks compared with patients who received the same chemotherapy every two weeks. The study demonstrated that 2-week intervals were more effective than 3-week intervals without increasing toxicity.[52]

This study led to VDC/IE being the standard of care in the United States. The chemotherapy is started before the local therapy and then continued postoperatively if there is no evidence of progression.

In the non-metastatic setting, induction with alternating VDC/IE is delivered in 6 cycles over 12 weeks. Following induction, local therapy is introduced during the 13 to 14 weeks. If surgery is planned, chemotherapy is restarted as soon as possible, usually within 1 to 2 weeks postoperatively. If radiotherapy is used, chemotherapy is typically delivered concurrently, except for anthracyclines—which are held during this period. Systemic treatment is typically concluded at weeks 42 to 48, depending on the protocol.[53]

Metastatic Disease

Patients who present with de novo metastatic disease have a generally poor prognosis; this constitutes approximately 25% of all patients with Ewing sarcoma.[54] The 5-year overall survival of this group is approximately 30%.[55] However, certain subsets of patients may have a better prognosis. Risk stratification systems based on location and overall disease burden have been proposed.[55] For example, patients with isolated pulmonary metastasis have a 5-year overall survival of 30% to 40% compared to patients with diffuse involvement of multiple organs, where the expected overall survival is 5% to 10% over the same time frame.[55] Table 3 below illustrates a risk stratification system taken from Khanna, et al, with modifications. Such a system may allow for escalation or de-escalation of treatment based on clinical features.

Table 3. Risk Stratification of Metastatic Disease

Similar systemic chemotherapy discussed previously are also used for metastatic disease. The addition of ifosfamide or etoposide to standard chemotherapy was found ineffective, with no survival benefit for patients with metastatic disease.[49][50]

A prospective study for the treatment of metastatic Ewing sarcoma evaluated the dose intensification approach (doxorubicin, vincristine with or without high-dose cyclophosphamide, followed by ifosfamide and etoposide); however, this approach did not change the survival rates.[56]

The role of high-dose chemotherapy with hematopoietic cell rescue (HDC/HSCT) was also studied for patients with metastatic disease. EURO-EWING 99-R2pulm randomized trial of patients with lung metastases demonstrated no significant survival benefits in patients who underwent busulfan/melphalan high-dose chemotherapy with autologous stem cell rescue (BuMel) compared with those who underwent conventional chemotherapy with whole-lung irradiation.

Recurrent Disease

Patients who relapse after definitive treatment have poor outcomes, with a 5-year survival of 13%.[57] The time to recurrence is considered a predictor for post-recurrence survival, with patients >/= 2 years having a 5-year survival of 30% versus those <2 years at 7%.[57] While most relapses occur within the first 2 years of treatment, long-term relapses can occur.[58] The cumulative risk of recurrence at 20 years is 13%.[59] 

The rEECur trial is a multiarm Phase II/III trial that investigated the use of several agents, including topotecan and cyclophosphamide, irinotecan and temozolomide, gemcitabine and docetaxel, or high-dose ifosfamide. This trial demonstrated the superiority of high-dose ifosfamide in prolonging survival, with a median survival of 16.8 months. Other systemic agents, such as eribulin in combination with irinotecan and nab-paclitaxel with gemcitabine, are underway (NCT03245450) (NCT02945800). Targeted molecular agents have also been of interest. Cabozantinib, a tyrosine kinase inhibitor, has demonstrated antitumor activity in heavily pre-treated patients.[60] Ganitumab, an anti-IGFR antibody, combined with palbociclib, a CDK4 inhibitor, produced disappointing results (NCT04129151).[3] Other agents targeting the EWS-FL11 fusion protein, such as TK216, are under investigation (NCT02657005).[3] Immunotherapy agents targeting PDL-1 with pembrolizumab or nivolumab have not produced tumor responses.[3] 

The lack of response may result from low PDL expression (19% of all tumor cell samples), limited tumor-infiltrating lymphocytes, and a low tumor mutational burden.[3] High-dose chemotherapy with autologous stem cell transplantation (HDT-ASCT) and no current prospective clinical trials are in progress.[3] This is an intensive systemic regimen that only a select number of candidates with relapse may be eligible to receive. Single-institution experience using HDT-ASCT demonstrated a 10-year overall survival of 46%, including relapsed and metastatic Ewing sarcoma. Patients with early relapse <2 years have an especially poor OS.[61]

Toxicity

Ewing sarcoma is associated with an increased risk of adverse health outcomes. Patients treated with current therapies are likely to develop severe neutropenia, which may be complicated by recurrent fever, opportunistic infections, and mucositis, despite the administration of granulocyte colony-stimulating factor. Potential adverse events of radiotherapy and chemotherapy include an increased risk of a second malignant neoplasm. The cumulative incidence of the second neoplasm in cancer survivors of large series was less than 2%, which usually occurs within 3 years of initial diagnosis.[62] 

The most common malignancies are acute myeloid leukemia, myelodysplastic syndrome, and sarcomas within the radiation field. Drugs that can cause secondary leukemias include alkylating agents (ie, cyclophosphamide), topoisomerase II inhibitors (ie, etoposide, teniposide), and anthracycline agents (ie, doxorubicin). However, the IESS trial comparing VAC with VAC-IE showed that adding etoposide was not associated with the increased risk of a second malignancy.[63] The C-arm of the INT-0091 study demonstrated that very high cumulative doses of ifosfamide and cyclophosphamide were associated with a 10% incidence of therapy-related leukemia. Cyclophosphamide and ifosfamide are associated with infertility. Also, ifosfamide can cause renal failure, slowly progressive chronic renal failure, or renal tubule cell dysfunction.

Anthracyclines, including doxorubicin, are well known to cause cumulative-dose-related cardiotoxicity. Treatment with doxorubicin can be complicated by cardiomyopathy in a dose-dependent manner and can manifest many years after the treatment. Prolonged infusion or dexrazoxane can be administered before doxorubicin to decrease the risk of cardiotoxicity.[64]

Staging

The commonly used staging system for Ewing sarcoma developed by the Musculoskeletal Tumor Society classifies Ewing sarcoma by grade (low grade being stage I, high-grade stage II, distant metastasis stage III) and compartmental status (located in the bone cortex versus extended beyond the bone cortex). Another classification method is TNM by the American Joint Committee on Cancer, which is based on tumor size, lymph node metastasis, distant metastasis, and tumor grade (cellular differentiation, mitotic rate, and extent of necrosis).

Prognosis

Distal extremity involvement in Ewing sarcoma is associated with a more favorable prognosis, than patients with lesions in proximal extremities.[64] Metastatic disease presence crucially impacts prognosis assessment.

Patients with solitary pulmonary metastases are more likely to have a better prognosis than patients with extrapulmonary metastatic sites, and unilateral lung involvement correlates with a better prognosis than bilateral lung involvement.[65] The development of both bone and lung metastases and extensive tumors are associated with poor prognosis. Tumor size is an important prognostic factor in studies. Data show that being under 15 years is another significant clinical prognostic factor. Clinical studies have shown that patients with minimal or no viable residual tumor at surgery following neoadjuvant chemotherapy did seem to have a better outcome than the patients with more significant amounts of viable tumors.[65] Additionally, inadequate response to neoadjuvant chemotherapy is associated with an increased risk of recurrence. EWSR1-ETS translocation is no longer considered an adverse prognostic factor.

Complications

Ewing sarcoma may be complicated by metastases, local recurrence, secondary malignancies, pathological fractures, surgery, and radiation-associated and chemotherapy-associated morbidities. Please refer to each subsection for complications specific to the treatment modality.