18F-FET is a fluorinated amino acid used in imaging CNS glial tumours. 18F-FET is taken up into neoplastic cells due to their increased amino acid uptake through an L-type amino acid transport system. Unlike GBCAs on MRI, uptake of amino acid tracers is independent of blood-brain-barrier integrity. FET PET has utility in the differential diagnosis between malignant and non-neoplastic lesions10 and to stratify between low-grade and high-grade gliomas. FET PET for surgical planning offers potential to undertake supramarginal resections of non-enhancing disease on MRI, assessment of residual disease on postoperative imaging3, and for biopsy planning to identify targetable higher grade foci (Figure 1).
FET-PET also offers an avenue to improve radiation therapy planning in patients with gliomas with morphometrically variable RT fields and dose escalation protocols. FET-PET is also useful to evaluate treatment response and on follow up imaging to improve the distinction between glioma recurrence and post-treatment changes.4
Diagnosis and Workup of Glioma:
18F-FET uptake is significantly higher in high-grade gliomas compared with low-grade gliomas6, 7. A recent meta-analysis demonstrated that 18F-FET PET has higher sensitivity compared to 18F-FDG PET in differentiating between high-grade and low-grade gliomas but with lower specificity; diagnostic performance values were similar to those of 11C-methionine PET.2 Dunet et al report a pooled analysis of 5 studies (119 patients) demonstrating that 18F-FET had superior performance to 18F-FDG and should be preferred when assessing a new isolated brain tumour.8 For glioma grading, however, both tracers showed similar performances. Notably, Dunet et al used the ratio of maximum uptake in tumour to background tissue (TBRmax) of 2.1 as a threshold for determining glioma versus alternative diagnosis, where 18F-FET demonstrated a 65% sensitivity and 56% specificity for diagnosis. Within our group, a retrospective 18F-FET study of 22 patients with a definitive glioma diagnosis, a TBRmax of 2.1 resulted in a sensitivity of 86.7% and specificity of only 37.5% and the study concluded that management of lesions with a threshold TBRmax of > 2.5 should be suspicious for glioma and warrant biopsy consideration whereas lesions with TBRmax > 2.1 should be closely monitored.9 For indeterminate brain lesions on MRI, our group (Chan et al) retrospectively studied 35 patients showing that 18F-FET demonstrated a high positive predictive value for glioma. Confirmatory trials are needed to establish the potential value of FET PET in guiding surgical management in this cohort of patients with indeterminate lesions on conventional MRI.10
FET-PET may better delineate brain-tumour extent/infiltration than conventional MRI. As FET uptake is independent of the blood-brain barrier, it has the particular advantage in delineating disease which is ‘non-enhancing’ on MRI with GBCAs. Fusing PET and MRI can overcome some of the inherent limitations of PET/CT spatial resolution when evaluating brain tumours compared to each individual imaging method alone. PET/MR offers the opportunity to combine these modalities.
Radiation Therapy Planning:
Our group continues to examine the role of FET-PET in radiotherapy for gliomas. In a prospective study of 10 patients with anaplastic glioma (AG), FET-PET was shown as a useful adjunct in radiation treatment planning and can potentially deliver a morphologically variable RT field with improved targeted dose to the tumour, whilst sparing normal brain. At our institution, we prospectively studied a novel approach for AG utilizing improved targeting of radiation therapy (intensity-modulated radiation therapy) to areas within the tumour at different dose levels (integrated boost) defined by MRI and nuclear medicine dose painting techniques.11(Figure 2).
In a retrospective study of 26 patients with glioblastoma multiforme, (as part of post-operative radiation therapy planning) a volumetric analysis between MRI derived clinical target volumes (CTV) and 18F-FET derived biological tumour volumes demonstrated a statistically significant difference between quantitative radiation planning maps. Whereby FET-PET improved delineation of GBM in cases with a suspected non-enhancing component and reduces the risk of potential geographical miss.12 Additionally, in patients with contraindication to MRI, FET-PET we have shown value in aiding radiotherapy planning.13
Surveillance Imaging:
The RANO working groups have primarily focused on MRI, however, newer consensus guidelines acknowledge management decisions should incorporate amino-acid PET for follow up of high-grade gliomas. Despite the increasing adoption of advanced MR imaging, including DSC/DCE quantitative perfusion and permeability analysis and ADC histography, uncertainty remains in many follow up scans. Difficulties implementing standardized MR protocols, a range of quantitative “thresholds” and frequent problems with MR artefact amplify the challenges when relying solely on MRI.
Our work demonstrating the fundamental utility of FET-PET in surveillance imaging has shown utility monitoring gliomas (particularly IDH-mutant gliomas) for response to chemotherapy (Figure 3) and for evidence of transformation or dedifferentiation (Figure 4).
Differentiating progression vs pseudo-progression and delayed post-therapy changes remains the fundamental challenge in high-grade glioma imaging surveillance. Recent literature demonstrates the high diagnostic accuracy in IDH-wildtype gliomas for differentiation of both pseudo-progression and delayed radiation necrosis vs true progression. A meta-analysis of 7 studies with a total of 172 lesions that underwent 18F-FET PET demonstrated a pooled sensitivity of 90% (CI 81-95%) and specificity of 85% (CI 71-93%) with regards to diagnostic accuracy in the differentiation of tumour progression from treatment-related changes, significantly outperforming 18F-FDG PET.1 Furthermore, 18F-FET images of glioblastoma patients with pseudo-progression show a slightly lower and more homogenous FET uptake, whereas patients with early tumour progression showed a more heterogeneous FET uptake and are reflected by the identified radiomics parameters.14 Dynamic amino acid uptake curves have been used by our group and others to further risk stratify regions of FET uptake and improve clinician confidence in conservative surveillance of some patients (Figure 5). Novel therapies including bevacizumab and features of ‘pseudo-response’ add another layer of diagnostic confusion on MRI surveillance in situations where FET-PET may have a role.
Limitations:
Care must be taken in the interpretation of FET-PET. FET as a tracer has inherent limitations – it often has low-grade background uptake in normal brain, can be seen in non-neoplastic pathology (ranging from non-neoplastic lesions, CVA to vascular anomalies). Importantly, although the sensitivity of FET-PET is high in IDH-wildtype lesions, lower grade (usually IDH-mutant gliomas) often do not show any FET uptake. Interpretation of FET-PET should be made in a multidisciplinary environment with both neuroradiologist and nuclear medicine input together with clinical colleagues.
Clinical characteristics between studies may differ significantly between institutional cohorts: IDH status, extent of surgery, delivered radiotherapy dose, follow-up time, percentage of patients with recurrence, etc. Consequently, performing external validation of quantitative radiomic parameters is challenging. Future validation in large prospective cohorts with similar clinical characteristics is required. Logistically, several factors beyond the diagnostic performance may influence the choice of a specific PET tracer for evaluating gliomas. These include cost, tracer availability, radiation dose, and the presence of a cyclotron facility, as well as safety, legal, organizational, and economic aspects.