The proposed histopathologic grading, to evaluate the effect of chemotherapy on the primary tumor, had the strongest correlation to clinical outcome. This method could therefore be used to identify patients with a high risk of recurrent disease. These patients could be randomized to receive alternative postoperative treatments to investigate whether more aggressive therapies will improve outcome.

Background. Neoadjuvant chemotherapy is the most accepted treatment for localized osteosarcoma. This has led to a great improvement in limb‐sparing surgery and in disease‐free survival. Patients with a good response to preoperative chemotherapy showed a higher disease‐free survival rate. Current studies examine the possibility of patients whose limbs could be rescued with a poor necrosis and a reduction of the side effects related to aggressive treatments. Methods. Between September 1986 and December 1989, 164 patients entered the second neoadjuvant study conducted at the Rizzoli Institute, Bologna, Italy, for non‐metastatic osteosarcoma of the extremities. Preoperative chemotherapy consisted of two cycles of high‐dose methotrexate intravenously (IV) followed by cisplatin intraarterially and doxorubicin IV. After surgery, patients classified as good responders (> 90% tumor necrosis) received three more cycles of these drugs, whereas poor responders (< 90% tumor necrosis) had more chemotherapy, which included ifosfamide and etoposide in addition to the other three drugs. Results. Limb salvage was performed in 83% of cases. At an average follow‐up of 54 months (36–76), 109 patients (66%) were continuously disease‐free, 2 died from doxorubicin cardiotoxicity, and 52 experienced metastases and 3 had local recurrence. In two of these three patients, metastases followed local recurrence. The 5‐year actuarial continuously disease‐free survival rate was 63%, with no differences between good and poor responders. Excluding 20 patients who had major protocol violations, the projected continuous disease‐free survival rate was 71%. Conclusions. With an aggressive neoadjuvant chemotherapy, it is possible to cure more than 60% of nonmetastatic osteosarcoma of the extremities, avoiding amputation in most cases. Ifosfamide and etoposide seem to be effective in patients who did not respond to preoperative chemotherapy.

Patients with metastatic osteosarcoma and localized chondroblastic osteosarcoma have a reduced chemosensitivity to primary chemotherapy with MTX, CDP, and ADM. MTX serum peak significantly influences tumor necrosis. A dose adaptation of MTX is recommended to obtain a serum peak of 700 micromol/L or greater when MTX is infused in 6 hours.

HDCT combined with surgery is feasible and can induce CR in a large portion of patients. Two points, however, need to be considered: only patients who are chemosensitive to induction treatment can obtain CR after HDCT, and the length of remission is short, because most patients relapse. Thus novel strategies are needed to maintain the remission status or to treat patients who do not respond to induction treatment.