Although survival of osteosarcoma patient has markedly improved, cases of non-responders to chemotherapy and late-relapsers are still perplexing. Our strategy was to analyze the impact of chemotherapy and surgery on the survival for each stage, and to evaluate long-term survival and find prognostic factors within the same stage.

From May 1985 to Feb. 1999, 461 osteosarcomas were enlisted at our department and among them 348 cases were evaluable. There were 1 IB, 4 IIA, 302 IIB, and 41 IIIB. Two hundred and fifty-five (IIA/IIB:4/251) out of 348 cases followed our protocol of chemotherapy and surgery. Two hundred and ten cases had neoadjuvant chemotherapy and 45 had adjuvant only.

Eleven year event free survival of the neoadjuvant group was 56.7%. On univariate analysis for 210 neoadjuvant cases, age (<13yrs), type of surgery (amputation), type of chemotherapy (methotrexate, adriamycin, cisplatin), pathologic response (>90%), local recurrence, pathologic fracture, location and size were statistically significant. But multivariate one revealed age, type of surgery, local recurrence and pathologic response as useful factors. There were 12 local recurrences (5.7%) and 100 metastasis among 255 stage II and their average onset from treatment was 17.8 month. Survival after metastasis was 5.6% at 55 months and the aggressively treated group made gains in survival (P<0.0001). Survival of 41 stage III was 0% at 64 months and this group also had an advantage in survival through intensive chemotherapy and surgery for primary and metastatic lesions (p=0.04).

Long-term survival of the treated 255 stage II group was 55% at 14 years. For stage II: age, local recurrence and pathologic response were meaningful prognostic factors. Aggressive surgery and chemotherapy were necessary to improve the survival of stage III and stage II with late metastasis group.