A comparison of local control achieved by surgery and radiotherapy in patients with non-metastatic pelvic Ewing sarcoma enrolled on INT-0091

Abstract

Purpose/ObjectiveRetrospective data suggests that surgery provides better local control (LC) than radiotherapy for Ewing sarcomas. However, lesions that are deemed not resectable are frequently large and may therefore be more prone to local failure by either modality. Pelvic lesions are commonly not amenable to surgery without disabling morbidity and thus definitive radiotherapy is often employed. The purpose of this analysis of data from the randomized controlled trial INT-0091 (CCG-7881 and POG-8850), is to investigate the relationship between the type of local control modality used, surgery alone (S), radiation alone (RT) or both (S+RT), and subsequent risk for local failure or adverse event for patients with non-metastatic Ewing sarcoma of the pelvis.Materials/MethodsPatients ≤30 years old with Ewing sarcoma, PNET of bone, or primitive sarcoma of bone were randomized to receive 49 weeks of standard chemotherapy with doxorubicin, vincristine, cyclophosphamide, and dactinomycin, or experimental therapy with these four drugs alternating with courses of ifosfamide and etoposide. The local control modality (S, RT or S + RT) was chosen by the treating physicians and administered at week 12. Patients with positive margins after surgery received post-operative radiation therapy. The cumulative incidences (CIs) of local failure (LF) as any component of failure and event-free survival (EFS) were calculated. An event was defined as a disease recurrence, second malignant neoplasm (SMN) or death prior to disease progression or SMN. Additionally, the effect of local control modality (S, RT or S + RT) was assessed after adjusting for the size of tumor (≤8cm, 9+ cm) and randomized treatment regimen.Results93 of 518 eligible patients had localized primary tumors of the pelvis. Of these, 71 had tumor measurements in a least one dimension and sufficient data and details of their local control modality for analysis. 12 received surgery alone, 42 received radiation alone, and 17 received both. The larger tumors were significantly more likely to have been treated with both S and RT. 5-year EFS for these 71 patients with pelvic tumors was 49%. There was no difference in EFS by tumor size, LC modality, or treatment regimen. The 5-year CI for LF was 21% (15% - LF only, 6% - LF and DF). There was no significant difference in LF rates by LC modality (S, RT or S + RT) after adjusting for size of tumor and treatment arm. The 3-year CI of LF in the small tumors was 33% (S), 24% (RT), and 0% (S + RT), p = 0.6. The 3 years CI of LF for the large tumors was 17% (S), 24% (RT) and 15% (S + RT), p = 0.81.Conclusions Purpose/ObjectiveRetrospective data suggests that surgery provides better local control (LC) than radiotherapy for Ewing sarcomas. However, lesions that are deemed not resectable are frequently large and may therefore be more prone to local failure by either modality. Pelvic lesions are commonly not amenable to surgery without disabling morbidity and thus definitive radiotherapy is often employed. The purpose of this analysis of data from the randomized controlled trial INT-0091 (CCG-7881 and POG-8850), is to investigate the relationship between the type of local control modality used, surgery alone (S), radiation alone (RT) or both (S+RT), and subsequent risk for local failure or adverse event for patients with non-metastatic Ewing sarcoma of the pelvis. Retrospective data suggests that surgery provides better local control (LC) than radiotherapy for Ewing sarcomas. However, lesions that are deemed not resectable are frequently large and may therefore be more prone to local failure by either modality. Pelvic lesions are commonly not amenable to surgery without disabling morbidity and thus definitive radiotherapy is often employed. The purpose of this analysis of data from the randomized controlled trial INT-0091 (CCG-7881 and POG-8850), is to investigate the relationship between the type of local control modality used, surgery alone (S), radiation alone (RT) or both (S+RT), and subsequent risk for local failure or adverse event for patients with non-metastatic Ewing sarcoma of the pelvis. Materials/MethodsPatients ≤30 years old with Ewing sarcoma, PNET of bone, or primitive sarcoma of bone were randomized to receive 49 weeks of standard chemotherapy with doxorubicin, vincristine, cyclophosphamide, and dactinomycin, or experimental therapy with these four drugs alternating with courses of ifosfamide and etoposide. The local control modality (S, RT or S + RT) was chosen by the treating physicians and administered at week 12. Patients with positive margins after surgery received post-operative radiation therapy. The cumulative incidences (CIs) of local failure (LF) as any component of failure and event-free survival (EFS) were calculated. An event was defined as a disease recurrence, second malignant neoplasm (SMN) or death prior to disease progression or SMN. Additionally, the effect of local control modality (S, RT or S + RT) was assessed after adjusting for the size of tumor (≤8cm, 9+ cm) and randomized treatment regimen. Patients ≤30 years old with Ewing sarcoma, PNET of bone, or primitive sarcoma of bone were randomized to receive 49 weeks of standard chemotherapy with doxorubicin, vincristine, cyclophosphamide, and dactinomycin, or experimental therapy with these four drugs alternating with courses of ifosfamide and etoposide. The local control modality (S, RT or S + RT) was chosen by the treating physicians and administered at week 12. Patients with positive margins after surgery received post-operative radiation therapy. The cumulative incidences (CIs) of local failure (LF) as any component of failure and event-free survival (EFS) were calculated. An event was defined as a disease recurrence, second malignant neoplasm (SMN) or death prior to disease progression or SMN. Additionally, the effect of local control modality (S, RT or S + RT) was assessed after adjusting for the size of tumor (≤8cm, 9+ cm) and randomized treatment regimen. Results93 of 518 eligible patients had localized primary tumors of the pelvis. Of these, 71 had tumor measurements in a least one dimension and sufficient data and details of their local control modality for analysis. 12 received surgery alone, 42 received radiation alone, and 17 received both. The larger tumors were significantly more likely to have been treated with both S and RT. 5-year EFS for these 71 patients with pelvic tumors was 49%. There was no difference in EFS by tumor size, LC modality, or treatment regimen. The 5-year CI for LF was 21% (15% - LF only, 6% - LF and DF). There was no significant difference in LF rates by LC modality (S, RT or S + RT) after adjusting for size of tumor and treatment arm. The 3-year CI of LF in the small tumors was 33% (S), 24% (RT), and 0% (S + RT), p = 0.6. The 3 years CI of LF for the large tumors was 17% (S), 24% (RT) and 15% (S + RT), p = 0.81. 93 of 518 eligible patients had localized primary tumors of the pelvis. Of these, 71 had tumor measurements in a least one dimension and sufficient data and details of their local control modality for analysis. 12 received surgery alone, 42 received radiation alone, and 17 received both. The larger tumors were significantly more likely to have been treated with both S and RT. 5-year EFS for these 71 patients with pelvic tumors was 49%. There was no difference in EFS by tumor size, LC modality, or treatment regimen. The 5-year CI for LF was 21% (15% - LF only, 6% - LF and DF). There was no significant difference in LF rates by LC modality (S, RT or S + RT) after adjusting for size of tumor and treatment arm. The 3-year CI of LF in the small tumors was 33% (S), 24% (RT), and 0% (S + RT), p = 0.6. The 3 years CI of LF for the large tumors was 17% (S), 24% (RT) and 15% (S + RT), p = 0.81. Conclusions