Brain tumors have a relatively high incidence (>14/100000 people/year) and represent a major cause of death in the population. The direct and indirect costs of brain tumors are high in the developed countries (5.2 bn EUR/year in the EU; 4.46 bn USD/year in the US). A combination of recent advancements in molecular neuroimaging, with positron emission tomography (PET) in the first place, providing clinicians with an improved diagnostic and therapy follow-up efficacy, novel approaches in the field of neurosurgery (including neuronavigation, intraoperative control of the nervous function, tumor histology and volume), and developments in treatment strategies (including new chemotherapeutics and new targeted agents, immunotherapies, sophisticated irradiation protocols) has in the past years improved the survival of brain tumor patients. A major component of further improvements is related to advancements in the development of novel molecular imaging biomarkers for brain tumor detection, including new PET radiopharmacons with high specificity, sensitivity and diagnostic accuracy. Despite the fact that FDG is the "working horse" of brain tumor imaging with PET and well over 90 % of diagnostic imaging studies in neuro-oncology are made with FDG world-wide, due to its sub-optimal specificity and sensitivity the search for non-FDG brain tumor PET radiotracers has been intensifying during the past decade in order to improve the diagnostic sensitivity, specificity and accuracy of molecular imaging of brain tumors. The most promising non-FDG brain tumor radiotracers include radioactively labeled nucleoside and amino acid analogues, tracers of oxidative metabolism, fatty acid metabolism and hypoxia, as well as receptor ligands of various kinds. The most widely tested non-FDG radiotracers include [11C]methionine (MET), [18F]fluorothymidine (FLT), [18F]fluoroethyl-l-tyrosine (FET), [18F]fluoro-α-methyltyrosine (FMT), [18F]fluoromisonidazole (F-MISO), 6-[18F]fluoro-dihydroxy-l-phenylalanine (F-DOPA), [11C]choline (CHO) and [18F]choline. The selective advantages of these radiotracers, compared to FDG, are varying, MET and FET appearing to be the most useful dedicated glioma radiotracers. Nevertheless, several other non-metabolic radiopharmaceuticals are also being tested or are in the validation phase. Although novel dedicated radiotracer candidates should offer an increased selectivity, specificity and diagnostic accuracy when compared to the recently existing brain tumor tracers, a dual or a multitracer approach may still offer the optimal solution in brain tumor imaging with PET in the near future.