The glucose consumption in tumoursin vivo as reflected by uptake of [18F]2-fluoro-2-deoxy-D-glucose (18FDG) using positron emission tomography (PET) is currently under investigation as a measure of tumour response to radiotherapy. The calculation of cerebral metabolic rate of glucose from18FDG-PET data requires a proportionality factor referred to as the lumped constant. In the presentin vitro study, the utilizations of18FDG and glucose have been measured in a human glioblastoma cell line (86HG-39) as a function of γ-radiation dose with various post-irradiation times and of different fractionation modes. The ratio of utilization of18FDG to that of glucose (RF/G), assumed to correspond to the lumped constant, was observed to increase 12 and 24 h after single fraction γ-exposure by factors ranging from 1.2 to 1.5 compared with the non-irradiated controls. It decreased after multiple fraction γ-exposure (4 × 2 Gy) by a factor of 0.7 compared with the single fraction schedule (1 × 8 Gy). The results suggest that the affinities of glucose transporters or hexokiriase enzyme or both for18FDG and glucose could be influenced by γ-irradiation in this tumour cell linein vitro. Apparent changes of the glucose consumption determined with PET in human tumours following radiotherapy may, therefore, not be solely due to changes in cellular metabolism or cell number but may also be due to changes in RF/G.