Purpose: To examine prognostic factors for primary site control and analyze late effects according to local treatment modality in non-metastatic Ewing sarcoma (ES).
Materials and methods: From 1976 to 2001, 76 patients with localized ES and a median age of 14.5 years were seen and treated at one institution. Tumors were located in the extremity in 38, pelvis in 13, spine in 11, trunk in 8, and head and neck in 6. Tumor size was < or = 8 cm in 44 and >8 cm in 32. Local therapy included radiotherapy (RT) alone in 40, surgery (S) alone in 27, and surgery followed by postoperative radiotherapy (S + RT) in 9. Chemotherapy (CT) was delivered to 65 patients (86%). Median follow-up for surviving patients was 9.7 years.
Results: The 5- and 10-year overall survival rates were 57.5% and 52.1% while the 5- and 10-year local control rate was 79.2%. The 5- and 10-year local control rates were 78.2% for RT, 77.6% for surgery, and 88.9% for S + RT (P = 0.68). Multivariate analysis showed that only the use of CT was found to be a prognostic factor for local control (P = 0.014). The 5- and 10-year local control rates were 83.7% for those receiving CT and 51.1% for those not receiving CT. For patients followed at least 5 years from diagnosis, late effects were seen in 10 of 19 (52.6%) receiving RT, 2 of 5 (40%) receiving S + RT, and 4 of 16 (25%) receiving surgery alone. The most common late effects with RT were muscular atrophy, limb length growth delay, and development of second malignancy. Scoliosis and decrease range of motion of an extremity were seen regardless of local treatment modality. Three patients who had surgery alone had an amputation whereas two who had RT had an amputation secondary to relapse or development of osteosarcoma.
Conclusion: In this single institution study, the use of CT was the only factor found to impact on local control. Late effects were common regardless of local control strategy.