BACKGROUND: High-grade bone osteosarcoma has a high relapse rate. The best treatment of local recurrence (LR) is under discussion. The aim of this study is to analyze LR patterns and factors prognostic for survival. METHODS: LR diagnostic modality (clinical or imaging), pattern of recurrence, and post-LR survival (PLRS) were assessed. RESULTS: Sixty-two patients were identified, with median age 21 years (range, 9-75 years), including 11 (18%) ≤15 years, 30 (48%) from 16 to 29 years; 21 (34%) were older. Patterns of relapse were LR only 58%, LR + distant metastases (DM) 42%. Seventy-nine percent of patients relapsed within 24 months, and diagnosis was clinical in 88%. LR treatment was surgery 85%, chemotherapy 55%, chemotherapy + surgery 45%. Surgical complete remission after LR (CR2) was achieved in 60% (LR 86%; LR + DM 23%). With a median follow-up of 43 months (range, 5-235 months), the five-year PLRS was 37%, significantly better for patients with longer LR-free interval (LRFI; ≤24 months 31% vs > 24 months 61.5%, P = 0.03), absence of DM (no DM 56% vs DM 11.5%, P = 0.0001), and achievement of CR2 (no CR2 0% vs CR2 58.5%, P = 0.0001). No difference was found according to age and chemotherapy (LR only: five-year PLRS: 53% without chemotherapy vs 58% with chemotherapy, P = 0.9; LR + DM: five-year PLRS: 25% without chemotherapy vs 9% with chemotherapy, P = 0.7). CONCLUSIONS: Early relapse is detected by symptoms in 90% of cases and associated with worse outcome. The achievement of CR2, not age, is crucial for survival. For patients with LR only, better survival was demonstrated, as compared with DM, and no improvement with chemotherapy after surgery was found.
BACKGROUND: High-grade bone osteosarcoma has a high relapse rate. The best treatment of local recurrence (LR) is under discussion. The aim of this study is to analyze LR patterns and factors prognostic for survival. METHODS:LR diagnostic modality (clinical or imaging), pattern of recurrence, and post-LR survival (PLRS) were assessed. RESULTS: Sixty-two patients were identified, with median age 21 years (range, 9-75 years), including 11 (18%) ≤15 years, 30 (48%) from 16 to 29 years; 21 (34%) were older. Patterns of relapse were LR only 58%, LR + distant metastases (DM) 42%. Seventy-nine percent of patients relapsed within 24 months, and diagnosis was clinical in 88%. LR treatment was surgery 85%, chemotherapy 55%, chemotherapy + surgery 45%. Surgical complete remission after LR (CR2) was achieved in 60% (LR 86%; LR + DM 23%). With a median follow-up of 43 months (range, 5-235 months), the five-year PLRS was 37%, significantly better for patients with longer LR-free interval (LRFI; ≤24 months 31% vs > 24 months 61.5%, P = 0.03), absence of DM (no DM 56% vs DM 11.5%, P = 0.0001), and achievement of CR2 (no CR2 0% vs CR2 58.5%, P = 0.0001). No difference was found according to age and chemotherapy (LR only: five-year PLRS: 53% without chemotherapy vs 58% with chemotherapy, P = 0.9; LR + DM: five-year PLRS: 25% without chemotherapy vs 9% with chemotherapy, P = 0.7). CONCLUSIONS: Early relapse is detected by symptoms in 90% of cases and associated with worse outcome. The achievement of CR2, not age, is crucial for survival. For patients with LR only, better survival was demonstrated, as compared with DM, and no improvement with chemotherapy after surgery was found.