[¹⁸F]fluoroethyl-tyrosine (FET) is an artificial amino acid taken up into upregulated tumoral cells by Na+ independent transport via the LAT system, independently of BBB leakage [31]. FET is not incorporated into proteins, as with the natural amino acids, and its uptake grade is not directly proportional to tumor differentiation status [27,32,33]. FET is diffuse in Europe and has been shown to provide high sensitivity and specificity for glioma detection and low uptake in the inflammatory and healthy brain [31,34], resulting in a reliable diagnostic tool for differentiating tumor recurrence/progression from TRC. Dynamic FET PET and time–activity curves (TACs) offer additional information on tracer kinetics. As known, in HGG, FET uptake is characterized by an early peak 10–15 min after injection, followed by a decrease in radiopharmaceutical’s uptake, similarly to recurrence [35,36]; differently, an LGG shows a typical delayed and steadily increasing tracer uptake similar to TRC [37]. These patterns are usually observed for FET PET and not for other amino acid tracers such as MET and FDOPA. Static PET scan protocols might not reveal the active metabolic tumor and might suffer from a lack of standardized acquisition protocols; thanks to the proprieties of not being metabolized after the entry into the cell, advanced pharmacokinetic analysis of TACs from dynamic FET PET scans, using compartment models, was exploited [34].

Several studies investigated the role of FET in the evaluation of recurrence and a variable diagnostic accuracy ranging between 81% and 99% was reported [38,39]. This wide range could be explained considering the different PET parameters analyzed, e.g., tumor-to-background ratios (TBR_max and TBR_mean), acquisition time protocols (static vs. dynamic), the uptake kinetics (time to peak—TTP, in minutes from the beginning of the dynamic acquisition up to the maximum SUV of the lesion) as well as different patient populations, tumor subtypes, and treatments [40]. Notably, guidelines reported a common threshold to assess glioma recurrence on FET imaging defining a TBR_mean of 2.0 and a TTP < 45 min [16]. Using these cutoff values, Galldiks and colleagues yielded the best result for identifying tumor recurrence or progression, with a sensitivity of 93%, a specificity of 100%, and an accuracy of 93% [35]. Nevertheless, PET parameters and the relative threshold are not standardized in all studies and different diagnostic performances were reached using different values. For example, in a cohort of 26 patients with GBM, a cutoff value of 1.9 for TBR_max allowed differentiating between true progression and late PSP. Moreover, in the same study, the dynamic acquisition also identified different curve patterns for differential diagnosis. A FET uptake peaking at a midway point (>20–40 min) followed by a plateau or a small descent was described as pattern II; conversely, the uptake peaking early, at 20 min, followed by a constant descent was named pattern III. Curve patterns type II and III resulted in being predictive for true progression and TRC, respectively [40]. However, it is important to underline that, in clinical routine, dynamic FET PET imaging has several limitations. As known, it requires longer acquisition times (50–60 min vs. 10–20 min for a static scan), which reduces patient compliance and could cause motion artifacts, with increasing costs of the investigation [35]. Recently, Bashir et al. performed a static FET PET study in a homogenous population of 146 suspected recurrent GBM patients 20 min after administration, demonstrating that FET parameters were significantly higher in patients with recurrence compared with patients with TRC (TBR_max, 3.2 vs. 1.6; TBR_mean, 2.0 vs. 1.6; biological tumor volume—BTV,14.8 cm³ vs. 0.01 cm³; p < 0.0001). Using a threshold of 2.0 for TBR_max, PET-based classifications of recurrent GBM or TRC were confirmed in 98.8% of patients [39]. In contrast to the reported higher performance of early FET imaging (10–20 min) for the primary diagnosis of glioma, in this setting of recurrent disease, static PET imaging acquired from 30 to 40 min demonstrated a sensitivity and specificity of 80.0% and 84.6%, respectively, at an optimal cutoff of TBR 2.07, providing slightly better discrimination than early images [36]. Disease progression assessment could also be challenging in recurrent GBM treated with bevacizumab, which is an antiangiogenic agent. Two clinical studies of patients with recurrent HGG undergoing bevacizumab therapy showed an advantage of FET PET over MRI in the early detection of tumor progression. Response to therapy as detected with PET (defined as >45% decrease in tumor volume) was also associated with longer overall and progression-free survival (OS and PFS) [41,42,43]. Although the potential role of FET PET in differentiating TRC from recurrence is well established, the methodology has yet to be standardized to define imaging protocols, as well as both the tumor and the normal brain reference regions, since the differentiation of a viable tumor from TRC is predominantly established by TBR [16]. In a 2019 study, the diagnostic performance of several analytic approaches in the setting of PSP in GBM was evaluated. All TBRs’ measures were significantly higher in patients with true tumor progression as compared with late PSP, regardless of the semiquantitative approach applied. Although these results are encouraging and the significance is promising, the need for a consistent method of background activity assessment is requested. A crescent-shaped background volume of interest (VOI), as a reproducible approach for methodological standardization, and an isocontour approach (including multiple voxels with the highest radiopharmaceutical uptake) were proposed to reduce noise, increase reproducibility, and avoid potential pitfalls of reference region definition (e.g., the inclusion of structural changes due to atrophy, trauma, or ischemia) [44,45].