2021 Background: Re-irradiation is an important treatment option for recurrent glioblastoma (rGBM). Here, we aimed to compare the oncological outcome of a re-irradiation target volume delineation based on O-(2-[18F]fluoroethyl)-L-tyrosine (FET) positron emission tomography (PET) with the one of a treatment based on contrast enhanced T1-weighted magnetic resonance imaging (T1Gd-MRI). Methods: GLIAA was a prospective, multicenter, randomized trial (NOA 10/ARO 2013-1, DKTK-a.). Patients with rGBM recurrence of 1-6 cm following initial radiotherapy were recruited at 14 centers in Germany and randomized 1:1 between a FET-PET- (arm A) and a T1Gd-MRI-based target volume delineation (arm B). The treatment consisted in a high-precision stereotactic re-irradiation with 39 Gy in 13 fractions. The primary endpoint was progression-free survival (PFS) from randomization. Secondary endpoints included overall survival (OS), local tumor control, and toxicity. Results: From November 26, 2013, to September 2, 2021, 271 patients were assessed for eligibility and 200 were randomized. 98 patients in the FET-PET arm and 97 patients in the T1Gd-MRI arm were treated per protocol. Median PFS was 4.0 months (95% confidence interval [CI] 3.7-5.2) in the FET-PET arm and 4.9 months (95% CI 3.7-6.0) in the GdT1-MRI arm (one-sided stratified log-rank test p=0.98; adjusted HR for the experimental versus the control arm 1.14 [95% CI 0.85-1.52], p=0.39). Median OS was 9.4 months (95% CI 7.8-11.1) in arm A and 9.0 months (95% CI 7.6-10.5) in arm B (HR 1.01 [95% CI 0.75-1.37], p=0.92). Local control rate at 12 months was 22% in the FET-PET arm (95% CI 14%-31%) and 20% in the GdT1-MRI arm (95% CI 12%-29%). Radiation necrosis as adverse event was documented in 25.5% of cases in arm A and in 21.6% in arm B. Of the 239 patients who received the FET tracer, 3 reported 5 severe adverse events in the timespan of 7 days after PET. No event was related to the application of the tracer. Conclusions: In this trial, FET-PET-based re-irradiation of rGBM was not superior to the GdT1-MRI-based treatment. Consequently, both options remain valid in this setting. Stereotactic re-irradiation with 39 Gy in 3 Gy fractions was shown to be safe and the FET-PET investigation was well tolerated in all cases. Due to the inclusion of FET-PET-positive patients only, our results do not impact the known diagnostic role of FET-PET in differentiating rGBM progression from post-therapeutic changes. Clinical trial information: NCT01252459 .