Background. The aim of this study was to investigate the potential of dynamic O-(2-[ 18 F]fluoroethyl)-L-tyrosine ( 18 F-FET) PET for differentiating local recurrent brain metastasis from radiation injury after radiotherapy since contrastenhanced MRI often remains inconclusive. Methods. Sixty-two patients (mean age, 55 ± 11 y) with single or multiple contrast-enhancing brain lesions (n = 76) on MRI after radiotherapy of brain metastases (predominantly stereotactic radiosurgery) were investigated with dynamic 18 F-FET PET. Maximum and mean tumor-to-brain ratios (TBR max , TBR mean ) of 18 F-FET uptake were determined (20-40 min postinjection) as well as tracer uptake kinetics (ie, time-to-peak and slope of time-activity curves). Diagnoses were confirmed histologically (34%; 26 lesions in 25 patients) or by clinical follow-up (66%; 50 lesions in 37 patients). Diagnostic accuracies of PET parameters for the correct identification of recurrent brain metastasis were evaluated by receiver-operating-characteristic analyses or the chi-square test. Results. TBRs were significantly higher in recurrent metastases (n = 36) than in radiation injuries (n = 40) (TBR max 3.3 ± 1.0 vs 2.2 ± 0.4, P < .001; TBR mean 2.2 ± 0.4 vs 1.7 ± 0.3, P < .001). The highest accuracy (88%) for diagnosing local recurrent metastasis could be obtained with TBRs in combination with the slope of time-activity curves (P < .001). Conclusions. The results of this study confirm previous preliminary observations that the combined evaluation of the TBRs of 18 F-FET uptake and the slope of time-activity curves can differentiate local brain metastasis recurrence from radiation-induced changes with high accuracy. 18 F-FET PET may thus contribute significantly to the management of patients with brain metastases.