Benefits of 18F-FET PET in preoperative assessment of glioma heterogeneity demonstrated in two case reports

Gliomas are intra-axial brain tumors with high biological heterogeneity. MRI is still considered the gold standard of diagnostic algorithms. Especially in gliomas with atypical enhancement patterns, where foci with different grading can coexist, PET could increase sensitivity and specificity of pre-surgery imaging by monitoring processes such as cell proliferation, membrane biosynthesis, and glucose consumption [1]. Radiolabeled amino acids are currently in general use as PET tracers due to their higher signal-to-noise ratio as compared to the traditional tracer 18F-2-fluoro-2-deoxy-D-glucose (18F-FDG). In contrast to glucose derivatives, the uptake of amino acids in macrophages and other inflammatory cells is lower, which makes amino acid tracers more specific for tumor detection [2]. Moreover, amino acids are accumulated in malignantly transformed cells owing to increased expression of type L amino acid transporters in tumor vasculature [3]. Additionally, the countertransport system A is overexpressed in neoplastic cells and seems to correlate positively with tumor cell growth rate [4]. Elevated transport of amino acids is, therefore, not only a result of increased protein synthesis but also reflects the increased demand by different types of metabolism in the tumor cell [5]. Currently, [18F]-fluoro-ethyl-L-tyrosine (18F-FET) is one of the most promising amino acid radiotracers in the diagnostic of low-grade glioma and its possible hot spots. Besides diffusion through the impaired blood-brain barrier (BBB), 18F-FET can additionally cross the BBB via endothelial L-transporters [6]. It also retains the relatively long half-life of 109 min for fluor-18, in contrast to, e.g., carbon-11 labeled methionine with a half-life of 20 min. 18F-FET can be produced in large amounts for clinical purposes and is applicable for PET studies in a satellite concept similar to the widely used 18F-FDG. In contrast with 18F-FET, 3′-deoxy-3′-18F-fluorothymidine (18F-FLT) is a radiolabeled analog of the DNA nucleoside that enters cells by active transport through nucleoside transporters, and by passive diffusion [6]. As it does not cross the intact BBB [7] and reflects tissue proliferation rate [8], the diagnostic yield of 18F-FLT in low-grade gliomas (LGG) is questionable because of their more indolent biological course.

Two cases of suspected LGG are presented below, where advanced PET imaging with 18F-FLT and 18F-FET was performed in addition to conventional MRI. The aim of this project to analyze which of the compared radiotracers is more suitable to be used in a standard protocol of glioma evaluation.

A 43-year-old woman suffering from complex partial epileptic seizures was examined by a neurologist. An acute CT revealed a tumor with calcifications in the left parietal lobe. MRI showed a brain lesion with no enhancement after contrast injection (Fig. 1). The patient was referred to a neurosurgeon and advanced PET methods were indicated to investigate possible tumor heterogeneity. 18F-FLT PET images were acquired 10 minutes after tracer injection in a single static image. 18F-FET PET images were acquired as dynamic scanning between 5 minutes and 45 minutes after injection, each image acquisition lasting 5 min. Tracer accumulation was evaluated in all images to show tracer kinetics. 18F-FLT PET showed moderately increased accumulation of the radiotracer in a small portion of the tumor (maximal standardized uptake value [SUV_max] 1.75, contralateral reference 0.4) as well as intensive positivity of the surrounding parietal bone. This finding supported the diagnosis of LGG but was in contrast with the result of dynamic 18F-FET PET, where regions with slowly increasing tracer activity representing LGG component were found together with regions of tracer kinetics more typical for high-grade glioma (HGG), i.e., an early tracer activity peak followed by its gradual decrease (Figure 1). The maximal SUV_max was detected 10 minutes after tracer injection in the postcentral gyrus (SUV_max 15.3). Increased tracer uptake was also noticed in the supramarginal gyrus (SUV_max 13.4). Different dynamics were detected in the periventricular area of the left occipital horn, where mildly elevated values (mean value of SUV_max for 5^th to 40^th min was 6.4) remained stable during the whole scanning time. Contralateral supramarginal gyrus was used as a reference with mean SUV_max 2.5. The afore mentioned occipital area could correspond to a low-grade portion of the tumor. The patient underwent 5-ALA navigated surgery with intraoperative monitoring. We combined somatosensory evoked potential (SSEP) with the phase reversal and direct cortical (bipolar) / subcortical (monopolar) stimulation to monitor motor cortex and the corticospinal tract that enabled maximal safe resection. Throughout the resection, SSEP remained stable and subcortical motor stimulation indicated the cortico-spinal tract was at a safe distance. A small residue was left in the area adjacent to the basal ganglia due to the high risk of complex neurological deficit in the region of low SUV-max indicating low-grade component of possible oligodendroglial tumor. Histological examination confirmed the diagnosis of oligodendroglioma, WHO grade 2, IDH-mutant and 1p/19q-codeleted with focal upgrading into anaplastic oligodendroglioma, WHO grade 3, IDH-mutant and 1p/19q-codeleted. This corresponded with the different SUV_max curves in different tumor parts on 18F-FET PET. After recovery, the patient was referred to an oncology service for adequate comprehensive cancer treatment.

A 44-year-old man with headache and lower limb dysesthesias lasting for approximately 2 months was diagnosed (CT scan) with a large tumor of the left frontal lobe with vast calcifications. MRI showed a non-enhancing inhomogeneous lesion infiltrating the corpus callosum and basal ganglia, compressing the left ventricle and causing a midline brain shift of 18 mm (Fig. 2). Both 18F-FLT and 18F-FET PET scans were obtained in the same fashion as in the previous case. 18F-FLT PET showed a homogenous, mildly elevated accumulation of the tracer in the whole lesion (SUV_max 1.4, contralateral reference 0.6; values of SUV_max above 2.0–2.5 are usually considered significantly pathologic). Interestingly, 18F-FET PET showed pathologic accumulation of radiotracer in the central tumor area (SUV_max 9.9 after 5 min) and its caudal part (SUV_max 8.6 after 5 min). Moreover, the SUV curve showed an early peak and a slow decrease of accumulation with time (to SUV_max 6.4 in both of these areas), whereas the rest of the tumor had more stable dynamics. With knowledge of possible upgrading according to 18F-FET PET, the patient underwent gross-total resection. The final histopathological diagnosis was anaplastic oligodendroglioma, WHO grade 3, IDH-mutant and 1p/19q-codeleted, originating from oligodendroglioma, WHO grade 2. In this case, some samples were harbored topically, aiming the spots of high 18F-FET PET attenuation, and we detected a significant correlation between elevated radiotracer accumulation and histological foci of upgrading. This seems promising, however, further investigation on a bigger cohort is required. After surgery, oncologic treatment for HGG was indicated based on the histological findings.

In both cases, the Siemens Biograph 64 mCT Flow scanner was used with the Siemens syngo.via diagnostic software. Scanning was performed with the following parameters: for 18F-FLT PET a static 15-min scan was acquired from 10^th min after the tracer injection (standardly used protocol in our institution); for 18F-FET PET a dynamic acquisition was performed in 5-min bins from 5^th to 50^th min after tracer application. Contralateral area to the tumor site was used for normalization and comparison. For attenuation reconstruction purposes, the low-dose CT image was acquired, no contrast agent was used. Acquired images were subsequently fused with 3 T (tesla) MRI, T2 FLAIR (T2-weighted-fluid-attenuated inversion recovery) sequence and further evaluated as PET/MR fusion

18F-FLT is a suitable radiotracer in diagnostics of glioblastoma, WHO grade 4, since its uptake is highly related to BBB disruption and reflects tissue proliferation [10]. However, it is unable to detect the aggressive tissue transformation before the BBB is damaged. Therefore it has a limited diagnostic efficiency in LGGs where BBB usually remains intact [7]. It also has a high positivity in the bone marrow of the skull, causing an unsatisfying signal-to-noise ratio. Although 18F-FET is not yet a part of widely used diagnostic protocols in the Czech Republic, it has some advantages over 18F-FLT that makes it applicable in clinical practice. 18F-FET easily crosses the intact BBB using the combination of passive diffusion and active transport mechanisms and has a relatively long half-life that makes its clinical use more convenient. Lesions in 18F-FET PET images demonstrated higher tumor to background ratio than in 18F-FLT PET images. Also, the change of accumulation in time as shown in 18F-FET PET provided additional information compared to 18F-FLT PET. The intensity of 18F-FET accumulation measured by SUV_max correlates with amino acid uptake and protein synthesis. Our results are in accordance with an increasing number of studies demonstrating the correlation of 18F-FET PET findings obtained from static and dynamic acquisition to malignant tumor parts within suspected WHO grade II gliomas [7,8].

In conclusion, the results of PET/CT imaging in both case reports suggest the superiority of 18F-FET over 18F-FLT in revealing focal areas of possible upgrading in suspected LGG. According to the authors' experience with PET imaging, the use of 18F-FET PET in the routine diagnostic protocol of gliomas is recommendable, especially in case of suspected LGG to rule out possible upgrading.

The Editorial Board declares that the manu script met the ICMJE “uniform requirements” for biomedical papers.
Redakční rada potvrzuje, že rukopis práce splnil ICMJE kritéria pro publikace zasílané do biomedicínských časopisů.

Prof. Radim Jančálek, MD, PhD
Department of Neurosurgery
St. Anne’s University Hospital
Pekařská 53, 656 91 Brno
Czech Republic
e-mail: radim.jancalek@fnusa.cz
Accepted for review: 5. 1. 2021
Accepted for print: 29. 7. 2021