Neoadjuvant chemotherapy for nonmetastatic osteosarcoma of the extremities.

Between September 1986 and December 1988, 125 patients with osteosarcoma of the extremities entered the second neoadjuvant study at the authors' institution. Patients received preoperatively two cycles of methotrexate (MTX) intravenously, followed by cisplatinum (CDP) intraarterially, plus adriamycin (ADM) intravenously. After surgery, the patients classified as "good responders" (more than 90% tumor necrosis) received ADM, MTX, and CDP, while the "poor responders" (less than 90% tumor necrosis) had a longer chemotherapy that included ifosfamide and etoposide (VP-16) in addition to MTX, ADM, and CDP. Limb salvage was possible in 85% of patients, 8% had an amputation, and 7% had a rotationplasty. The surgical margins were adequate (radical or wide) in 88% of cases and inadequate (marginal or intralesional) in 12%. At an average follow-up period of 28 months (range, 13 to 41), 109 patients (87%) remained continuously disease free, 15 (12%) relapsed with pulmonary metastases, and one patient (0.8%) had a local recurrence. Compared with the first neoadjuvant study at the authors' institution that used only MTX and CDP preoperatively, the percentage of limb salvages, "good responders," and continuously disease-free survival at two years was significantly higher in the second Rizzoli neoadjuvant study (85%, 74%, and 87% versus 77%, 52%, and 59%). Systemic toxicity because of chemotherapy was superimposable. A retrospective analysis of the real dose intensity for each patient demonstrated a correlation between the intensity of chemotherapy and prognosis.