Individual Risk Evaluation for Local Recurrence and Distant Metastasis in Ewing Sarcoma: A Multistate Model

Background: This study aims to investigate the effect of surgical margins, histological response and radiotherapy, on local recurrence (LR), distant metastasis (DM) and survival in Ewing sarcoma patients. Methods: We performed a retrospective analysis of 982 Ewing sarcoma patients who underwent surgery after induction chemotherapy. To study disease evolution a multistate model with initial state surgery, intermediate states LR, pulmonary metastasis (DMpulm), other DM±LR (DMother) and final state death was estimated. The effect of risk factors on different disease states was estimated using Cox proportional hazard models. Hazard ratios (HRs) along with 95% confidence interval (95%CI) are provided. Overall survival (OS) from surgery was calculated using Kaplan-Meier's estimates. Findings: The median follow-up was 7·6 years (95%CI 7·2-8·0). 5-year OS was 74% (71-77) for localized disease, 56% (47-55) for pulmonary metastasis, 43% (33-53%) for extrapulmonary metastasis. Risk factors for LR are pelvic location, HR 2·04(95%CI 1·10-3·80), and marginal/intralesional resection, HR 2·28(1·25-4·16), administration of radiotherapy is protective, HR 0·52(0·28-0·95), for all tumor sites combined; risk factors for DMpulm are <90% necrosis, HR 2·13(1·13-4·00), and previous pulmonary metastasis, HR 4·90(2·28-8·52). Risk factors for DMother are 90-99% necrosis, HR 1·56(1·09-2·23), <90% necrosis, HR 2·66 (1·87-3·79), and previous bone/other metastasis, HR 3·08(2·03-4·70). Risk factors for death without LR/DM are pulmonary metastasis, HR 8·08(4·01-16·29), bone/other metastasis, HR 10·23(4·90-21·36) and <90% necrosis. Early LR (0-24 months) negatively influences survival, HR 3·79(1·34-10·76). Once DMpulm/DMother arise only previous bone/other metastasis remain prognostic for death, HR 1·74(1·10-2·75). Interpretation: Disease extent at diagnosis and histological response are the main risk factors for progression to distant metastasis or death. Tumor site and surgical margins are important risk factors for LR. In case disease progression occurs, previous risk factors lose their relevance, and the balance between toxicity of intensive salvage treatments and quality of life in the remaining life span of these patients should be carefully considered. In case of isolated LR, time to recurrence is important for decision-making, since early LR (0-24 months) negatively influences survival. Radiotherapy seems to be protective for LR especially in pelvic/axial locations. The low percentage of LR in extremity tumors and associated toxicity questions the need for radiotherapy in extremity Ewing sarcoma. Funding Statement: This study was supported by a donation from the Team Westland Foundation (to PDSD), a German Cancer Aid grant 108128 to UD; ERA-Net-TRANSCAN consortium, project number 01KT1310, and Euro Ewing Consortium EEC, project number EU-FP7 602856, both funded under the European Commission Seventh Framework Program FP7-HEALTH (to UD); and by David Dressler+ (to UD). Declaration of Interests: The authors have no conflict of interest. Ethics Approval Statement: This retrospective study was reviewed and approved by the institutional review board of Leiden University Medical Center (Leiden, The Netherlands) and a waiver for informed consent was granted.