A study was performed with the aim of investigating some of the methodological factors affecting the ability of quantitative 2-[18F]-fluoro-2-deoxy-d-glucose (FDG) positron emission tomography to assess tumour malignancy. Twenty-nine patients with soft tissue masses were studied using a 6-hour scanning protocol and various indices of glucose metabolism were compared with histological grade. Significant differences were observed in the time-activity response of benign and high-grade tumours. High-grade sarcomas were found to reach a peak activity concentration approximately 4 h after injection whereas benign lesions reached a maximum within 30 min. This translated to improved differentiation between these two tumour types using a standard uptake value (SUV) derived from images acquired at later times. An SUV measured 4 h post-injection was found to be as useful an index of tumour malignancy as the metabolic rate of FDG determined using either Patlak or non-linear regression techniques. Each of these indices had a sensitivity and specificity of 100% and 76% respectively for the discrimination of high-grade sarcomas from benign tumours.