Treatment algorithm for locally recurrent osteosarcoma based on local disease-free interval and the presence of lung metastasis

BACKGROUND. Local recurrence in osteosarcoma is clinically distinct from metastasis, although associated with a similar reduction in survival. The prognostic factors in locally recurrent osteosarcoma were investigated and these factors were translated into a management strategy. METHODS. In all, 407 consecutive patients with skeletal osteosarcoma between 1977 and 2002 were analyzed. Twenty-three patients with resectable local recurrence were analyzed. Clinical and tumor-related factors were assessed for significance in relation to survival and a management strategy was formulated based on factors found to be independently significant for survival. RESULTS. Seventeen of the 23 patients underwent primary resections and initial treatment, yielding an overall local recurrence rate of 4.2% for resectable cancer. Median time to local recurrence was 13 months (95% confidence interval, 9-16 months). The 5-year and 10-year survival rates in the recurrent cases were 29% and 10%, respectively. All patients received chemotherapy both for their primary and recurrent disease. Increased risk of local recurrence (P <.0001) was strongly correlated with positive margins of resection. The rate of local recurrence was not related to chemotherapy-associated necrosis in the primary tumor. Nevertheless, neoadjuvant therapy halved the risk of local recurrence (odds ratio, 1.92; P =.3, power 10%). The strongest correlate with poor survival was local recurrence within the first year after primary resection (P =.001), followed by metastasis at the time of first local recurrence (P =.04) and failure to achieve clinical remission after disease recurrence (P =.04). Chemotherapy-associated necrosis and margins of resection of the primary tumor were not significant prognostic variables for survival. Survival differed significantly among patients defined by local disease-free interval and lung metastasis (P =.0001). They required an individualized approach as captured in the management algorithm. CONCLUSIONS. There is a residual risk of local recurrence in patients despite favorable chemotherapy-associated necrosis and negative margins of resection. A treatment strategy emphasizing clinical remission at all identifiable sites offers the highest likelihood of survival in this patient population.