Key points are not available for this paper at this time.
PURPOSE: To define prognostic factors for response and long-term outcome for a wide spectrum of osteosarcomas, extending well beyond those of the typical young patient with seemingly localized extremity disease. PATIENTS AND METHODS: A total of 1,702 consecutive newly diagnosed patients with high-grade osteosarcoma of the trunk or limbs registered into the neoadjuvant studies of the Cooperative Osteosarcoma Study Group before July 1998 were entered into an analysis of demographic, tumor-related, and treatment-related variables, response, and survival. The intended therapeutic strategy included preoperative and postoperative chemotherapy with multiple agents as well as surgery of all operable lesions. RESULTS: Axial tumor site, male sex, and a long history of symptoms were associated with poor response to chemotherapy in univariate and multivariate analysis. Actuarial 10-year overall and event-free survival rates were 59.8% and 48.9%. Among the variables assessable at diagnosis, patient age (actuarial 10-year survival ≥ 40, 41.6%; < 40, 60.2%; P = .012), tumor site (axial, 29.2%; limb, 61.7%; P < .0001), and primary metastases (yes, 26.7%; no, 64.4%; P < .0001), and for extremity osteosarcomas, also size (≥ one third, 52.5%; < one third, 66.7%; P < .0001) and location within the limb (proximal, 49.3%; other, 63.9%; P < .0001), had significant influence on outcome. Two additional important prognostic factors were treatment related: response to chemotherapy (poor, 47.2%; good, 73.4%; P < .0001) and the extent of surgery (incomplete, 14.6%; macroscopically complete, 64.8%; P < .0001). All factors except age maintained their significance in multivariate testing, with surgical remission and histologic response emerging as the key prognostic factors. CONCLUSION: Tumor site and size, primary metastases, response to chemotherapy, and surgical remission are of independent prognostic value in osteosarcoma.
Bielack et al. (Fri,) studied this question.