Radiation-induced osteosarcomas (R-OS) have historically been high-grade, locally invasive tumors with a poor prognosis. The purpose of this study was to perform a comprehensive literature review and analysis of reported cases dealing with R-OS in the pediatric population to identify the characteristics, prognostic factors, optimal treatment modalities, and overall survival of these patients.

A MEDLINE/PubMed search of articles written in the English language dealing with OSs occurring after radiotherapy (RT) in the pediatric population yielded 30 studies from 1981 to 2004. Eligibility criteria included patients <21 years of age at the diagnosis of the primary cancer, cases satisfying the modified Cahan criteria, and information on treatment outcome. Factors analyzed included the type of primary cancer treated with RT, the radiation dose and beam energy, the latency period between RT and the development of R-OS, and the treatment, follow-up, and final outcome of R-OS.

The series included 109 patients with a median age at the diagnosis of primary cancer of 6 years (range, 0.08-21 years). The most common tumors treated with RT were Ewing's sarcoma (23.9%), rhabdomyosarcoma (17.4%), retinoblastoma (12.8%), Hodgkin's disease (9.2%), brain tumor (8.3%), and Wilms' tumor (6.4%). The median radiation dose was 47 Gy (range, 15-145 Gy). The median latency period from RT to the development of R-OS was 100 months (range, 36-636 months). The median follow-up after diagnosis of R-OS was 18 months (1-172 months). The 3- and 5-year cause-specific survival rate was 43.6% and 42.2%, respectively, and the 3- and 5-year overall survival rate was 41.7% and 40.2%, respectively. Variables, including age at RT, primary site, type of tumor treated with RT, total radiation dose, and latency period did not have a significant effect on survival. The 5-year cause-specific and overall survival rate for patients who received treatment for R-OS involving chemotherapy alone, surgery alone, and surgery plus chemotherapy was 17.3% and 17.3%, 56.6% and 50.3%, and 71.0% and 68.3%, respectively (p < 0.0001, log-rank test).

The type of treatment for R-OS was the most significant factor for cause-specific and overall survival. Patients who develop R-OS should be aggressively treated, because the outcome is not as dismal as once thought.