10515 Background: After the introduction of pre-operative chemotherapy in the early 1980s, treatment of osteosarcoma had not advantages from new drugs/modalities. Aim of this work is to analyze improvements in overall survival for patients treated over 21 years (1982–2002), with a 5-year minimum follow-up, in the largest series from a single institute ever reported, including all high grade osteosarcomas, despite histology varieties, age, site, and stage. Data are also analyzed in subgroups to define patients who benefited most. Methods: All diagnoses of high grade osteosarcoma were included. Of the 1,656 consecutive cases observed, 198 patients were excluded (41 consultation only, 129 low-grade varieties and 28 lost to follow-up). Within 1,456 included patients, 1,032 had characteristics to be enrolled in conventional clinical trials (classic histology, age < 41, localized and extremity disease). Results: Considering all patients, with a median follow-up of 12 years (5–25 yrs), 754 (51.7%) are alive, 613 continuously disease-free. Survival at 5, 10, and 15 years is 57%, 52%, and 51% respectively. Patients candidates for clinical trials have a survival rate of 68%, 64%, and 61% respectively. Survival for the other patients is 30%, 25%, and 24% respectively. Jointpoint Statistical Analysis at real 5-year follow-up shows a yearly statistically significant improvement in survival of 1.31% (95% CI 0.5–2.1), from 51% for patients treated in 1982 to 68% for those treated in 2002. Within the subgroups, survival statistically improved in patients candidates to protocols, those who relapsed, or presented with metastatic disease at diagnosis, or had axial tumors. Surgery was also analyzed, with a statistical significant increase in the percentage of limb salvage procedures without an increased rate of local recurrences. Conclusions: Despite the lack of new drugs for osteosarcoma, survival has statistically improved, especially for those patients with the worst outcome. Aggressive treatments are therefore recommended for all patients including those with poor prognosis. No significant financial relationships to disclose.