To evaluate PET/CT in pediatric primary bone tumours (PBT), the accuracy, clinical impact, prognostic indicators in predicting tumour response to therapy and determining epiphyseal involvement were compared to CI. Methods: A retrospective review of PET/CT scans with CI was performed. Lesions were compared to a reference standard: histopathology or follow up >6mths. Pt based analysis was performed for clinical impact. Prognostic indicators (SUVmax, tumour size) were compared to histopathology response post chemotherapy. Results: 43 pts (av age12.9 yrs) with osteosarcoma (18), Ewing’s sarcoma (21), PNE (4) were analysed. 109 PET/CT scans with CI scans were evaluated (371 lesions). 33 lesions were discordant. Accuracy of PET/CT was higher for all lesions than CI (95% vs92%) but sensitivity was lower (79% vs 83%). Excluding lung lesions, sensitivities increased for PET/CT and CI (92% vs 89%). 9pts had PET/CT staging and follow up with histopathological evaluation post chemotherapy: 2pts poor responders, 7 good responders. Good responders had a higher SUVmax at diagnosis compared to poor responders (av13.84 vs 7.95) but reduced more [10.5(70%) vs 3.5(45%)]following chemotherapy. There were no false negatives for epiphyseal involvement for PET/CT and CI but one PET/CT was false positive. Conclusion: PET/CT is less sensitive in small lung lesions, but more sensitive in other areas compared to CI. SUVmax at diagnosis is a poor predictor of response, but percent decrease post therapy was associated with therapeutic response. Change in tumour size on MR is a poor predictor of response. There is improved clinical impact with PET/CT in patient management.