PURPOSE: Especially in Europe, amino acid PET is increasingly integrated into multidisciplinary neuro-oncological tumor boards (MNTBs) to overcome diagnostic uncertainties such as treatment-related changes. We evaluated the accuracy of MNTB decisions that included the O -(2-[ 18 F]-fluoroethyl)-L-tyrosine (FET) PET information compared with FET PET results alone to differentiate tumor relapse from treatment-related changes. PATIENTS AND METHODS: In a single academic center, we retrospectively evaluated 180 MNTB decisions of 151 patients with CNS WHO grade 3 or 4 gliomas (n = 122) or brain metastases (n = 29) presenting equivocal MRI findings following anticancer treatment. All patients underwent FET PET imaging besides MRI before MNTB discussion. Additionally, the patient's clinical status and pretreatment were considered for decision-making. The diagnostic performance was calculated for FET PET findings alone and MNTB decisions that included FET PET results using 2 × 2 contingency tables. MNTB decisions were validated using the neuropathological result in 43% (n = 78) or clinicoradiologically in 57% (n = 102). RESULTS: FET PET results alone yielded an accuracy of 87% (sensitivity, 90%
specificity, 65%
positive predictive value, 95%). When integrating FET PET results for decision-making in the MNTB setting, the accuracy increased to 95% (sensitivity, 99%
specificity, 70%
positive predictive value, 96%
P = 0.002). In MNTB decisions concerning glioblastoma patients, the median survival was 2.4 times longer when FET PET suggested treatment-related changes (15.6 vs 6.4 months
P = 0.009). CONCLUSIONS: Our results suggest that MNTB discussion further enhances the FET PET value for identifying brain tumor relapse. A prospective evaluation of FET PET results with and without integration in an MNTB is warranted.