Pseudoprogression (PsP) is characterized by therapy-associated but not tumor growth-associated increases of contrast-enhancing glioblastoma lesions on MRI. Although typically occurring during the first 3 months after radiochemotherapy, PsP may occur later in the course of the disease and may then be particularly difficult to distinguish from true tumor progression. We explored PET using O-(2-(18)Ffluoroethyl)-L-tyrosine ((18)F-FET-PET) to approach the diagnostic dilemma. Twenty-six patients with glioblastoma that presented with increasing contrast-enhancing lesions later than 3 months after completion of radiochemotherapy underwent (18)F-FET-PET. Maximum and mean tumor/brain ratios (TBRmax and TBRmean) of (18)F-FET uptake as well as time-to-peak (TTP) and patterns of the time-activity curves were determined. The final diagnosis of true progression versus late PsP was based on follow-up MRI using RANO criteria. Late PsP occurred in 7 patients with a median time from radiochemotherapy completion of 24 weeks while the remaining patients showed true tumor progression. TBRmax and TBRmean were significantly higher in patients with true progression than in patients with late PsP (TBRmax 2.4 ± 0.1 vs. 1.5 ± 0.2, P = 0.003; TBRmean 2.1 ± 0.1 vs. 1.5 ± 0.2, P = 0.012) whereas TTP was significantly shorter (mean TTP 25 ± 2 vs. 40 ± 2 min, P < 0.001). ROC analysis yielded an optimal cutoff value of 1.9 for TBRmax to differentiate between true progression and late PsP (sensitivity 84%, specificity 86%, accuracy 85%, P = 0.015). O-(2-(18)Ffluoroethyl)-L-tyrosine PET provides valuable information in assessing the elusive phenomenon of late PsP. Clin Cancer Res; 22(9); 2190-6. ©2015 AACR.