PET imaging of gliomas using novel tracers: a sleeping beauty waiting to be kissed

Figure 1. (A)¹⁸F-FDG; (B) contrast-enhanced MRI; (C) ¹¹C-MET PET. Glioblastoma in the right frontal lobe, which is hard to delineate in the ¹⁸F-FDG PET. However, amino-acid PET with ¹¹C-MET clearly shows the lesion with excellent tumor to background contrast.

Approximately 1–2% of all tumors are brain tumors. Gliomas represent 45% of all brain tumors (50% being highly malignant), followed by meningiomas (27%), with an incidence of 2–16/100,000 [1]. The American Cancer Society estimated that, in 2005, primary brain tumors were the cause of death of approximately 12,760 women and men [2,3]. Precise tumor localization and exact delineation from surrounding healthy brain tissue is of the utmost importance for planning tumor resection and/or radiation therapy, as well as for assessment of tumor recurrence after primary therapy. Moreover, novel targeted therapeutics mainly used after primary resection and radiation therapy lead to an increasing demand for functional and molecular imaging techniques for assessment of therapeutic response. MRI, at present, is still the gold standard in the primary diagnosis of brain tumors. Conventional morphological MRI provides information with respect to size and localization as well as additional insights on secondary phenomena, that is edema and bleeding, which cannot be supplied in the same quality by PET [4]. However, the specificity for the differentiation of vital tumor tissue versus inflammatory processes, edema and reactive changes, is limited. Moreover, tumor borders determined by the use of contrast-enhanced MRI often under- or over-estimate the true extent of the tumor [5,6]. It is our firm belief, that in this respect the molecular information of PET using novel tracers provides vital information in combination with classical neuroimaging techniques, such as MRI and computed tomography (CT). In addition, functional MRI techniques including diffusion-weighted imaging (DWI), magnetic resonance spectroscopy (MRS), diffusion-tensor imaging (DTI), dynamic contrast-enhanced imaging (DCE-MRI) and dynamic susceptibility imaging (DSC MRI) are in this context, in our opinion, not competing but complementary modalities to PET.

Amino acid-based PET tracers for brain tumor imaging

While PET and PET/CT using the tracer ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) has become a success story for many tumors outside the CNS, the unfavorable characteristics of ¹⁸F-FDG in the CNS have hampered its success for imaging of brain tumors. While brain tumors, like other tumors, may show an increased glucose metabolism, which can be visualized and quantified with ¹⁸F-FDG, the major disadvantage of ¹⁸F-FDG for the diagnosis of a brain tumor is the high background uptake caused by the high glucose metabolism of the brain itself, resulting in a low signal–background ratio. To improve PET imaging in the CNS, radioactively labeled amino acids have been synthesized with many fewer limitations compared with ¹⁸F-FDG, that is ¹¹C-methionine (MET), ¹¹C-tyrosine, ¹¹C-leucine and amino acid derivatives, such as ¹⁸F-fluorethyltyrosine (FET) (Figure 1). Most primary gliomas show a high uptake of amino acids in comparison with normal brain tissues resulting in an excellent signal–background ratio. The enhanced uptake of amino acids into brain tumor tissue seems to be mainly attributable not to an increase of transport systems, the L or L-type amino acid transporters (LAT) [7–9]. The advantage of the ¹⁸F-labeled tracer ¹⁸F-FET compared with ¹¹C-labeled amino acids is the longer half-life of ¹⁸F (110 min) compared with ¹¹C (20 min). Therefore, ¹⁸F-FET can be distributed to centers without an on-site cyclotron in a satellite concept comparable with ¹⁸F-FDG. Comparative studies of ¹¹C-MET and ¹⁸F-FET with respect to uptake and image contrast revealed similar results for both tracers [10–13]. The following main indications for PET neuroimaging with amino-acid tracers have evolved:

• • Pretreatment definition of tumor extent for preoperative staging, for radiation therapy planning and for selecting the most appropriate biopsy site [14–17];

• • Detection of tumor recurrence and differentiation of recurrence and radiation necrosis [18–20];

• • (Early) therapy response assessment [17,21–23].

Primary diagnosis & work-up of brain tumors

Owing to the low uptake of ¹⁸F-FET in normal brain tissue (gray and white matter) compared with the high uptake in many primary brain tumors, amino acid tracers allow for the delineation of tumor boundaries with high accuracy [24]. There is a significantly higher uptake in malignant lesions compared with chronic and subacute nontumor lesions, which may demonstrate similar contrast enhancement patterns such as malignant lesions in CT or MRI [25]. Even low-grade tumors show a significant tracer uptake in many cases [26]. A threshold has been established for the definition of a ‘pathological’ tracer uptake: a ratio of tumor uptake to the uptake of the contralateral hemisphere of 1.6 or 1.5 or larger for FET and MET, respectively, is considered typical for vital tumor tissue [25,27]. Using these thresholds, malignant lesions can be differentiated from benign lesions with a sensitivity and specificity of approximately 80%. However, 25% of low-grade gliomas do not take up amino acid tracers, limiting the value of amino acid tracers to rule out low-grade gliomas. Clinically speaking, a high ¹⁸F-FET uptake is highly suggestive for a primary brain tumor, however, it is not possible to differentiate between low- and high-grade tumors. No uptake virtually excludes high-grade gliomas but does not rule out low-grade tumors, which can be PET negative.

MRI-guided stereotactic biopsy might not always result in a correct tumor tissue biopsy: in some cases the changes in the MRI signal may exceed the true tumor boundaries or tumor may be present in an area with no contrast enhancement in case of higher-grade tumor areas [12,24,27,28]. By contrast, biopsy studies revealed true positive amino acid tracer uptake even in non-solid tumor parts, resulting in a better detection of infiltrative growth of gliomas [29,30]. Moreover, this leads to a better differentiation between tumor tissue and reactive changes, such as edema, compared with MRI or CT [16,31]. Additionally it improves the detection rate of residual tumor tissue after resection and results in a better definition of target volumes before radiation therapy [5,6,32].

Detection of tumor recurrence

Another limitation of MRI is the detection of recurrent disease. Often, residual tumor tissue cannot be accurately distinguished from postoperative changes or radiation necrosis [13]. By contrast, ¹⁸F-FET-PET and other amino acid tracers can verify tumor recurrence with a high sensitivity and specificity and also differentiate between post-therapeutic changes and tumor tissue. A sensitivity of 100% and a specificity of 93% has been reported for this indication [20].

Response assessment

Concerning evaluation of therapy response, several studies demonstrated promising results for ¹⁸F-FET-PET. Examinations during chemotherapy and radioimmunotherapy demonstrated a high correlation of ¹⁸F-FET-PET uptake and response to treatment [33,34]. Functional changes can be visualized earlier in ¹⁸F-FET-PET than morphological changes in MRI [21]. On the other hand, ¹⁸F-FDG has proven to be of limited value for early therapy response assessment [35,36].

Grading & prognosis

The value of amino acid tracers as imaging markers for prognosis is still controversial. Whereas some studies only see a limited value [37] others propose a high value, particularly for grade II gliomas [38–40]. Furthermore, recently published data showed a potential correlation of dynamic ¹⁸F-FET-PET imaging and clinical outcome of the patients [41]. The value of amino acid tracers for grading so far is considered to be rather limited owing to the fact that even low-grade gliomas eventually show high tracer uptake. However, some recent studies have suggested dynamic acquisition and evaluation of uptake kinetics may lead to better estimation of tumor grading [42,43].

The assessment of grading and prognosis may represent one of the few remaining indications for ¹⁸F-FDG. Glucose metabolism can be used to characterize the aggressiveness of brain tumors and may supplement pathological grading [44,45]. However, a differentiation between WHO grade III and IV tumors is not possible owing to the comparable high uptake. Studies comparing ¹⁸F-FDG uptake and patient survival demonstrated a high correlation in patients with primary gliomas and recurrence of gliomas [44,46,47].

Imaging of specific biological processes of brain tumors with novel PET tracers

While diagnosis of brain tumors and tumor recurrence is mainly the domain of amino acid tracers, several tracers may show a high potential to evaluate specific biological features of tumors outside and within the CNS but still require further validation.

Hypoxia tracers, such as ¹⁸F-FAZA or ¹⁸F-FMISO, can be used to evaluate radiation sensitivity of brain tumors. This is of major relevance as hypoxic tumor areas do not respond well to radiation therapy [1,48]. However, limited permeability of the blood–brain barrier may hamper their application in CNS.

For imaging of proliferative activity, ¹⁸F-FLT has shown promising results. ¹⁸F-FLT is metabolized by thymidine kinase 1 and, thus, is a marker for DNA synthesis. Its uptake correlates better with Ki-67 proliferation index compared with ¹⁸F-FDG [49]. In recurrent disease ¹⁸F-FLT performed better compared with ¹⁸F-FDG in detecting tumor tissue and was a better predictor for disease progression and survival. Moreover, ¹⁸F-FLT holds promise as an imaging biomarker for therapy response assessment [50,51].

Concerning molecular imaging of angiogenesis, imaging specific biological targets of the angiogenic cascade is promising [52]. In this respect the integrin α_vβ₃ is an interesting target, as it is expressed on activated endothelial cells and is known to be overexpressed in gliomas [53]. The feasibility of imaging α_vβ₃ expression in patients with tumors outside the CNS and in gliomas has been demonstrated with the tracer ¹⁸F-Galacto-RGD [54–57]. Future studies will have to show if the concept of molecular imaging of angiogenesis is of clinical relevance, for example for response assessment of antiangiogenic therapy.

Conclusions & perspective

While ¹⁸F-FDG-PET still plays a role for assessment of tumor grade and dedifferentiation of gliomas over time, amino acid tracers such as ¹⁸F-FET are vastly superior to ¹⁸F-FDG for most indications (e.g., tumor delineation, assessment of recurrence and therapy response). Interestingly the regional use of amino acid PET varies substantially. In Europe, especially in Germany, ¹⁸F-FET-PET is increasingly used for glioma imaging in centers where PET is available. By contrast, the USA amino acid PET is not yet frequently performed and ¹⁸F-FDG is still the standard tracer for primary brain tumors. Thus PET imaging of gliomas with amino acid tracers is a ‘sleeping beauty’ in many parts of the world, waiting to be awakened by a kiss! Hopefully, the combined use of PET using novel tracers and advanced MRI such as DCE-MRI, DWI and MRS will increase the benefit for patients in the future. In this respect, great hope lies in combined MR-PET imaging, already realized with the brain PET MR-insert [58]. With fully integrated MR-PET systems currently being developed, a one-stop-shop approach for a complete assessment of brain tumor biology and morphology within the concept of individualized medicine seems feasible in the not too distant future.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.