The aim of the present study was to compare the contribution of the labelled tracers [C-11]methionine (Met) and [F-18]-fluorodeoxyglucose (FDG) in positron emission tomography (PET)-guided stereotactic biopsy of non resectable brain lesions. Twenty-five patients underwent combined Met-PET-, FDG-PET- and computerized tomography (CT)- or magnetic resonance (MR)-guided stereotactic biopsy according to a previously described technique for stereotactic FDG-PET. Met-PET and FDG-PET images were analyzed to determine which tracer offers the best information to guide at least one stereotactic biopsy trajectory. Histological diagnosis was obtained in all patients (23 tumours and 2 non-tumorous lesions). All tumours had an area of abnormal Met uptake and were biopsied under PET-guidance. FDG uptake in the tumour was higher than in the grey matter and was used for target selection in 12 of 23 tumours. Eleven of them were located in the basal ganglia or the brainstem. Met was used for target selection in 11 of 23 tumours where there was no FDG uptake or where FDG uptake was equivalent to that of the grey matter. Ten of them were located in the cortex. Two nontumoral lesions had no Met uptake and were biopsied under CT- or MR-guidance only. Forty-three out of 53 stereotactic trajectories obtained in these 25 patients were based on PET-defined targets and had an area of abnormal Met uptake. These trajectories always yielded a diagnosis of tumour. Moreover, all tumorous trajectories had an area of abnormal Met uptake. Finally, all non-diagnostic trajectories (n = 4) were CT/MR-defined because there was no area of abnormal Met uptake. These results suggest that patients who can benefit the most from Met-PET guidance could be selected pre-operatively. In conclusion, this work shows that Met is a good alternative to FDG for target selection in PET-guided stereotactic brain biopsy.