Radiation Oncology/Sarcoma/Ewing's sarcoma

Ewing's Sarcoma Epidemiology Tumor of neuroectodermal origin, part of the Ewing Sarcoma Family of Tumors Usually presents in adolescents (40%), but 30% in <10 year olds Second most common bone tumor in children, after osteosarcoma Approximately 225 cases annually in U.S. Primarily in white children; black and Asian children rarely affected Clinical Presentation Localized pain and swelling Systemic symptoms (fever, low appetite, weight loss) in ~30% Distribution Lower extremity 40-45% (femur 22%, tibia 11%, fibula 9%) Pelvis 20-25% Chest wall 15-20% Upper extremity 10% (humerus 10%) Metastatic disease (20-25%) Primary spread is hematogenous Most commonly to lungs, bones, BM, soft tissue, brain, spine Bilateral bone marrow biopsy part of staging, regardless of tumor size Radiologic Findings Should include bone scan, CXR, CT or MRI of primary, CT of chest Plain films show "onion skinning" (soft tissue mass growing out from the bone, giving rise to multilamellated periosteal reaction) vs "sunburst" pattern seen in osteosarcoma. Staging Please see the bone tumors staging page Pathology Small round blue cell tumor, likely arising in the bone marrow Fusion between EWS gene and a partner gene which presumably dysregulates cell growth t(11;22) EWS-FLI1 (85%) t(21;22) EWS-ERF (10-15%) Other EWS fusions occur rarely (ETV1, E1AF, FEV) Prognostic factors Disease site: non-pelvic is favorable, with distal, ribs and other having the best prognosis. Pelvic site is unfavorable. Proximal sites are intermediate. Age: younger is favorable Size: >8cm is unfavorable Labs: anemia, elevated ESR, leukocytosis, and elevated LDH associated with worse prognosis Not prognostic: sex and time from onset of symptoms Treatment Overview Ewing's sarcoma is essentially an occult metastatic disease, and chemotherapy is the backbone of treatment Radiation alone had cure rate ~10%, with majority failing distally Chemotherapy is typically given for 12-15 weeks prior to local therapy Local control is imperative, either with surgery, or radiation therapy, or both No randomized studies comparing the two treatment approaches Surgery favored if complete resection is feasible without significant morbidity and functional loss Radiation favored for central lesions Radiation dose Doses >60 Gy result in unacceptable risk of secondary bone malignancies Doses <40 Gy have unacceptable local failures Currently, ~45 Gy are given for microscopic disease and ~55.8 Gy for gross disease (DeVita 8th ed) Whole lung radiation used for consolidation after chemotherapy Chemotherapy regimens Standard 5-drug U.S. regimen (VCD + IE): vincristine, doxorubicin, and cyclophosphamide, alternating with ifosfamide and etoposide x48 weeks Original IESS regimen (VAC): vincristine, actinomycin D, cyclophosphamide VAC + ADR - vincristine, actinomycin D, cyclophosphamide + adriamycin Radiation Localized disease St. Jude (1978-88) -- used chemo before surgery Prospective, non-randomized. 60 pts. Localized osseous Ewing's. Induction chemo (ACx5) -> Resection (if possible) -> RT with concurrent chemo -> more chemo (total 57 wks). RT to initial (pre-chemotherapy) osseous tumor and residual soft tissue tumor extension. 3 cm margins. No RT given if resection with negative margins. 35-41 Gy for positive margins. 35 Gy for those with CR after chemo. 50.4 Gy for those with residual tumor after chemo or a tumor >8 cm initially. 1.5-1.8 Gy/fx. Chemo consisted of VCR/DAC during RT, followed by CYC/ADR and VCR/DAC. 1938559, 1991 — "Ewing's sarcoma: local tumor control and patterns of failure following limited-volume radiation therapy." Arai Y et al. Int J Radiat Oncol Biol Phys. 1991 Nov;21(6):1501-8. 5-yr EFS 59%, similar to other studies. 64% achieved CR to induction chemo. LC 68% at 5-yrs. LC depended on tumor size, +/- 8 cm. POG 8346 (1983-88) Randomized. 178 pts. Osseous Ewing's only (excluded PNET and extraosseous). Induction chemotherapy: CYC/ADR x 5. Assess response: If PD, XRT and salvage chemo If CR/PR and tumor involved expendable bone (proximal fibular, distal 4/5 of clavicle, body of scapula, iliac wing, ribs), had surgery. Received post-op RT for microscopic or gross residual disease. If not receive surgery, received 55.8 Gy RT and randomized to whole bone RT (39.6 Gy + boost with 2 cm margin) vs a tailored, involved field (same as boost). During RT, received VAC chemo followed by weekly VCR. Maintenance chemo given for a total of 50 wks. 9747829, 1998 — "A multidisciplinary study investigating radiotherapy in Ewing's sarcoma: end results of POG #8346. Pediatric Oncology Group." Donaldson SS et al. Int J Radiat Oncol Biol Phys. 1998 Aug 1;42(1):125-35. 5-yr LC for resected tumors was 88% vs 65% with RT alone. No difference in LC between those randomized to whole bone vs involved field. LF was in-field in 62% vs outside in 24%. Conclusion: Most failures are systemic rather than local. Thus, interpretation of local failures is difficult due to competing risks. Disseminated disease EURO-EWING 99 (1998-2006) 19924786 -- "The value of local treatment in patients with primary, disseminated, multifocal Ewing sarcoma (PDMES)." (Haeusler J, Cancer. 2010 Jan 15;116(2):443-50.) Retrospective. 120 patients. Local treatment of the primary with surgery 22%, surgery + RT 17% or definitive RT 33%. Local treatment of mets surgery 5%, surgergy + RT 7%, definitive RT 27%. No local therapy in 27% Outcome: 3-year EFS 24%; by modality surgery 25% vs surgery + RT 47% vs RT 23% vs no local therapy 13% (SS). 3-year EFS if treatment of primary and met 39% vs either primary or met 17% vs no local therapy 14% (SS) Conclusion: Local therapy of involved sites important for patients with disseminated Ewing sarcoma and should complement systemic treatment whenever possible Intergroup IESS Trials IESS-III (1988-92) 518 pts. Ewings's, PNET of bone, or primitive sarcoma of bone. Allowed both localized and metastatic (23%). Randomized to VAC+ADR or VAC+ADR alternating with ifosfamide and etoposide. Chemotherapy was given q3weeks x 17 courses, total of 49 weeks. At week 12, radiation therapy or surgery was performed. RT was 45 Gy to initial tumor volume (with 3 cm margin) + 1080 cGy boost = 5580 cGy. For those undergoing resection, RT was to 45 Gy for microscopic residual (with 1 cm margin). 12594313, 2003 — "Addition of ifosfamide and etoposide to standard chemotherapy for Ewing's sarcoma and primitive neuroectodermal tumor of bone." Grier HE et al. N Engl J Med. 2003 Feb 20;348(8):694-701. For non-metastatic pts, 5-yr EFS 69% vs 54% for VAC+ADR+IE vs VAC+ADR; RR=1.6. 5-yr OS 72% vs 61%; RR=1.6. Greater reduction in LR than in distant mets. Greater benefit for large primary tumors or pelvic tumors. For pts with mets, no difference between regimens: 5-yr EFS 22% and OS 34%. Conclusion: improved survival with addition of ifosfamide and etoposide (in non-metastatic pts). IESS-II (1978-82) 214 pts. Localized, non-pelvic Ewing's sarcoma of bone. Randomized to VAC+ADR as high-dose intermittent regimen or as moderate-dose continuous regimen (as per IESS-I). RT given in both arms for all pts except those with resection with negative margins. Chemotherapy given q3weeks (high dose intermittent) or weekly (continuous). High-dose: VA-VC x 6, then VD-VC x 7. Continuous: VC x 6, then VD-VC4-VCA x 7. RT at beginning of chemo. 2099751, 1990 — "Multimodal therapy for the management of nonpelvic, localized Ewing's sarcoma of bone: intergroup study IESS-II." Burgert EO Jr et al. J Clin Oncol. 1990 Sep;8(9):1514-24. Median f/u 5.6 yrs. 5-yr DFS, RFS, and OS 68% vs 48%, 73% vs 56%, 77% vs 63% for high-dose vs continuous. LR in only 9%. Conclusion: improved survival with more aggressive chemotherapy. Intergroup Ewing's Sarcoma Study IESS-I (1972-1978) 342 pts. Localized Ewing's sarcoma of bone, previously untreated. 7 pts treated with amputation were not randomized but were entered on treatment 1. Initially group I institutions randomized to +/- VAC chemo. Changed protocol in 1973 because it was realized that adjuvant chemotherapy was necessary. Randomized 3:2 to 1) RT to primary plus VAC + Adriamycin or 2) RT plus VAC -- group I institutions Randomized 3:2 to 3) RT to primary plus VAC and bilateral pulmonary RT (BRP) or 2) RT plus VAC (same as above) -- group II institutions Chemotherapy given x 6 weeks. Vincristine and cyclophosphamide given weekly and adriamycin given with the last dose. After 6 weeks rest, pts had a 7 week course of continuation therapy that consisted of dactinomycin IV x 5 days followed 9 days later by VCR and cyclophosphamide weekly x 5 weeks. For treatment 1, adriamycin given with the last course in the 7th week of each course. RT treated the entire involved bone to 45-55 Gy (dose depends on age), followed by 10 Gy boost to gross radiographic tumor + soft tissue mass with margin. For pts receiving lung RT, 15-18 Gy given at 150-180 cGy/day. 7029293, 1981 — "Multimodal therapy for the management of primary, nonmetastatic Ewing's sarcoma of bone: an Intergroup Study." Nesbit ME Jr et al. Natl Cancer Inst Monogr. 1981 Apr;(56):255-62. 2213103, 1990 — "Multimodal therapy for the management of primary, nonmetastatic Ewing's sarcoma of bone: a long-term follow-up of the First Intergroup study." Nesbit ME Jr et al. J Clin Oncol. 1990 Oct;8(10):1664-74. Median f/u 6 yrs. 5-yr RFS treatment 1 - 60%, 2 - 24%, 3 - 44%. Similar trend for OS. Worse survival for pelvic sites. No evidence of difference among treatments for pelvic sites. 15% LR overall. DM in 30%, 72%, and 42%. BPR was not effective in preventing lung mets. Conclusion: improved survival with addition of Adriamycin to VAC. Risk of Second Malignancy St. Judes; 1996 8874344 -- "Second malignancies after Ewing's sarcoma: radiation dose-dependency of secondary sarcomas." (Kuttesch JF, J Clin Oncol. 1996 Oct;14(10):2818-25.) Retrospective. 266 patients with Ewing's sarcoma. Median F/U 9.5 years (3-30) 16 second malignancies (10 sarcomas, 6 others). Median latency 7.6 years (3.5 - 25.7) 20-year cumulative incidence: 9.2% for any malignancy, 6.5% for sarcoma Dose-response: none in <48 Gy, 130/10,000 for >60 Gy Conclusion: Overall risk of 2nd malignancies similar to other pediatric tumors. RT dose-dependency justifies modification in approach to reduce radiation doses