Abstract
BACKGROUND: We sought to identify the significant prognostic parameters of patients with osteosarcoma over the past three decades using a population-based registry. METHODS: A total of 3765 patients with osteosarcoma were identified in the SEER database. Information regarding patient demographics, clinical and treatment characteristics, cause of death and survival were extracted. Kaplan-Meier, Log-Rank, and Cox regression were used for analysis. RESULTS: On multivariate analysis only age group '<25 years', 'local' stage and 'low' grade, 'appendicular skeleton' and employment of 'surgical resection' showed a disease-specific survival benefit with a P value < 0.001. The long-term survival improved in the interval from 1973 to 1985 from approximately 55 to 65% but subsequent improvement has been limited only to patients with high-grade disease. CONCLUSION: When comparing survival rates by decade of diagnosis, it appears that improvement in survival since 1985 is limited to patients with high-grade disease only. LEVEL OF EVIDENCE: The level of evidence for this article is 2.