European Musculo-Skeletal Oncology Society Meeting, 4 & 5 May 2000, London: Abstracts and Poster Presentations and Nurses Symposium

s 107 mean age of 19 years (range 6–50). The mean follow-up was 39 months (range 7–120 months). The vast majority had clear resection margins. Histological response was category 1 in 27% of the patients, and category 2 in 73%. Clinical features, treatment, and prognosis were analysed. Survival rates were examined using Kaplan–Meier Analysis. Results: The average interval between the first resection and local recurrence was 15 months (range 2–109 months). Forty-one patients (76%) had local recurrence in deep soft tissue, seven in bone, and six in subcutaneous tissue. Twenty-six patients (49%) had lung metastasis at the time of local recurrence, whilst 21 patients (38%) developed it later. Thirty patients (57%) were treated with resection of the recurrent lesion and 18 (32%) were treated with amputation. One-, 3and 5-year survival rates after local recurrence were 0.57, 0.38, and 0.22, respectively. Conclusion: (1) 87% of patients with local recurrence developed metastases either concurrently or at a later date. Immediate amputation did achieve local tumour control. However, the survival rate was not statistically higher; (2) 87% of the local recurrence arise in soft tissue. Therefore, careful attention should be paid to secure the wide margin around biopsy tract, muscle insertion to the affected bone, and neurovascular bundle at the time of initial resection. Prognosis after recurrence of osteosarcoma. A COSS report on 547 patients with relapse following complete remission B. Kempf-Bielack1, D. Branscheid, S. Flege, M. Kevric, R. Maas, H. Jürgens, & S. Bielack for the Cooperative OsteoSarcoma Study Group COSS. (1Department of Pediatric Hematology/Oncology, University Muenster, Germany) Introduction: The COSS-database was used to evaluate the prognosis after osteosarcoma relapse and to determine whether prognostic factors influencing survival post-relapse might be identified. Patients and methods: Among all patients with previously untreated high-grade osteosarcoma of the extremity or trunk registered into a neoadjuvant COSS-study before 7/98, 547 patients relapsed after having previously achieved a complete surgical remission. These were evaluated for patientand tumor-related characteristics identifiable at time of relapse which might influence post-relapse survival. Results: The median interval from diagnosis of osteosarcoma to relapse was short at 1.5 (0.1–10.3) years. One hundred and nineteen relapses were detected in the first, 242 in the second, 108 in the third, 36 in the fourth, 18 in the fifth year, and only 24 later. Median follow-up post-relapse was 1.0 year for all patients and 1.8 (0–17.3) years for 174 survivors. Three hundred and seventy-three patients died, 359 of osteosarcoma, 11 of other and three of unknown causes. Actuarial survival at 2, 5, and 10 years postrelapse was 34.4, 20.6, and 15.1%. Survival post-relapse did not correlate significantly with patient age or sex, the former site or size of the primary tumor, whether primary metastases had been present, or with tumor response to first-line chemotherapy. The probability to survive increased with increasing time between diagnosis and relapse (5-year survival 9.2% if interval < 1.5 years; 32.1% if 3 1.5 years). Five-year survival was 30.8% for 34 patients with isolated local recurrences and 21.3% for 481 patients with metastases only, but only 3.5% for 32 patients with combined local and systemic recurrences. Metastases involved the lung in 444 (355 as only site of relapse: 5-year survival 24.3%), distant bones in 88 (44 bone only: 23.6%) and other sites in 54 patients (11 other sites only: 20.0%). Five-year survival was dismal at 4.8% for patients with involvement of more than one system (local/lung/ bone/other), compared to 24.5% if only one system was involved. Conclusion: Survival following osteosarcoma relapse remains poor. Risk factors pointing to a high relapse risk after frontline therapy are generally unsuitable to predict outcome following relapse. A short latency period between diagnosis and recurrence and involvement of multiple locations at relapse identify a population with particularly unsatisfactory results. Supported by Deutsche Krebshilfe Post relapse survival in 159 patients with osteosarcoma of the extremity S. Ferrari, A. Briccoli, S. Rimondini, M. Mercuri, P. Picci, A. Tienghi1, A. Brach Del Prever2, A. Comandone3 & G. Bacci (Dipartimento di Oncologia Muscolo-Scheletrica Istituto Ortopedico Rizzoli; 1Oncologia Medica Ravenna; 2Clinica Pediatrica Università di Torino; 3Oncologia Medica Ospedale, Gradenigo Torino, Italy) An analysis of prognostic factors for survival was performed in 159 patients with osteosarcoma of the extremity who relapsed from 1986 to 1997. They had received a combined treatment with surgery of the primary lesion and neoadjuvant chemotherapy delivered according to four protocols based on high-dose methotrexate, doxorubicin, cisplatin and standard dose ifosfamide. The median relapse-free interval (RFI) was 23.6 months (1.9–98.3 months). One hundred and twenty-three (77%) patients relapsed only with lung metastases (monolateral 77, bilateral 46; median number of lung nodules, 3), 36 (23%) patients had recurrence also in other sites; 111 (70%) patients were free of disease after adequate surgical treatment; 85 (53%) patients received a second line chemotherapy treatment, based on high-dose Ifosfamide (15 g/m2, 5-day continuous infusion) in 50 (59%) of them. The estimated 5-year survival after relapse (S) was 27.5%. Univariate analysis identified the site of relapse (lung 5-year S 32%, other 12%, p < 0.003), RFI ( £ 24 months 5-year S 13%, > 24 months 5-year S 46%, p < 0.001), the site of lung metastases (monolateral 5-year S 37%, bilateral 17%, p < 0.001) and the number of lung metastases ( £ 2 nodules 5-year S 57%, > 2 nodules 5-year S 10%, p < 0.0001) as predictive factors for S. Chemotherapy offered a significant advantage only in patients with > 2 pulmonary nodules ( with chemotherapy 5-year S 17 %, without chemotherapy 5-year S 0%, p = 0.002), whereas no significant differences were seen in patients with one/two pulmonary nodules (42 vs. 62%, p = 0.46). A longer survival time and a higher 5-year S was seen in patients with other than lung metastases who received chemotherapy (with chemotherapy 5-year S 22%, without chemotherapy 5-year S 0%, p = 0.002). In a multivariate analysis including only patients with lung metastases the number of nodules ( £ 2 nodules RR 0.3, 95% CI 0.15–0.49, p < 0.0001) and RFI ( £ 24 months RR 2.18, 95% CI 1.3–3.7, p < 0.005) were predictive factors for S. About 25% of patients with recurrent osteosarcoma can be rescued. Lung location, a > 24-month relapse-free interval, £ 2 pulmonary nodules are positive predictive factors which identify patients who can be treated with surgery only. More than two pulmonary nodules and other than lung location identify poor risk patients who can have benefit from chemotherapy. Radiation induced bone sarcoma: a review of the EMSOS database H. Patterson & J.S. Whelan on behalf of EMSOS Data were sought from six centres; 65 patients (31 male and 34 female) were identified who satisfied the criteria for the diagnosis of radiation induced bone sarcoma. The median age at diagnosis of the condition for which the patient received their radiotherapy was 20 years (range 3 months to 66 years), and the most frequent diagnoses were Ewing’s sarcoma (13 patients), giant cell tumour (eight patients), breast cancer (seven patients), Hodgkin’s lymphoma (five patients), retinoblastoma (four patients) and primary bone lymphoma (three patients). The median age at