Non-routine Tracers for PET Imaging of High-grade Glioma

Tracers Using ¹¹C

¹¹CMethylAMD3465. The C-X-C chemokine receptor type 4 (CXCR4) plays a role during progression of some tumours including HGG (8). The PET imaging of the radiolabeled selective CXCR4 receptor antagonist ¹¹Cmethyl-AMD3465 has been studied in rats bearing C6 HGG xenografts. PET demonstrated specific accumulation of the tracer in the tumour and other CXCR4-expressing organs, such as lymph node, liver, bone marrow and spleen, and elevated tumour-to-muscle and tumour-to-plasma ratios were found, suggesting that ¹¹C-methyl-AMD3465 may detect physiological CXCR4 expression in HGG (8).

¹¹C-(R)PK11195. Elevated expression of the 18-kDa mitochondrial translocator protein (TSPO) may correlate with disease progression and aggressive, invasive behaviour of a variety of solid tumours, including HGG, while being rare in normal tissues. TSPO can be imaged by PET using the selective radiotracer ¹¹C-(R)PK11195. The binding potential of ¹¹C-(R)PK11195 in HGG has been found to be significantly higher than in low-grade gliomas (LGG) in a clinical study (9) (Figure 1A). TSPO in gliomas was expressed predominantly by neoplastic cells and its expression correlated positively with binding potential of ¹¹C-(R)PK11195 in the tumours. Hence, PET with ¹¹C-(R)PK11195 in HGG may predominantly reflect TSPO expression, and patient candidates for TSPO-targeting therapies may by stratified by this procedure.

*11C-Methyl-D-cysteine (“c-DMCYS). The D-amino acid isomer of *11C-methyl-cysteine has been proposed as a PET tracer potentially useful for discriminating tumour from inflamed tissues in animal models. PET imaging of orthotopic HGG models with ¹¹C-methyl-D-cysteine have further shown a fair tumour to normal brain tissue (T/N) ratio (10).

Tracers Using ⁶²CU and ⁶⁴CU

⁶²CU-diacetyl-bis (N4-methylthiosemicarbazone) (⁶²CU-ATSM). Tumour hypoxia assessed by ⁶²CU-ATSM PET/CT correlates with diffusion capacity obtained by diffusion-weighted imaging and may be useful for grading gliomas (11). Maximum standardized uptake (SUV) and T/N values of ⁶²CU-ATSM were significantly higher in human GB than in normal or lower grade tumour tissues, while limited differences were observed in parallel studies from the same group with ¹⁸F-FDG (12). ⁶²CU-ATSM PET/CT may also provide diagnostic information to differentiate GB from other brain neoplasms such as CNS lymphomas (11).

⁶⁴Cu-1,4,7-Triazacyclononane-1,4,7-triacetic acid (NOTA)-AC133 monoclonal antibody (⁶⁴Cu-NOTA-AC133 MAB). The AC133 epitope of CD133 is a widely used tumour stem cell marker for intra- and extracranial tumours. The non-invasive detection of AC133-overexpressing tumour cells by PET in subcutaneous and orthotopic glioma xenografts using antibody-based tracers has been described (Figure 1B) (13). Micro-PET with ⁶⁴Cu-NOTA-AC133 MAB yielded images with elevated T/N contrast, delineating subcutaneous tumour stem cell-derived xenografts. Intracerebral tumours as small as 2-3 mm were also discernible and the micro-PET images reflected the invasive growth pattern of orthotopic GIC-derived tumours with physiological density of AC133⁺ cells (13).

⁶⁴Cu-1,4,7,10-Tetra-azocyclododecane-N,N’N”,N”’-tetraacetic acid (DOTA)-TNYL-RAW peptide. The ephrin type-B receptor 4 (EPHB4) receptor, a member of the tyrosine kinase receptor family, may be overexpressed in both tumour cells and angiogenic blood vessels of HGG and the EPHB4-binding peptide TNYL-RAW labelled with ⁶⁴Cu has been examined as a PET tracer for orthotopic human HGG (14). TNYL-RAW was conjugated to the radiometal chelator DOTA and the conjugate was then labelled with ⁶⁴Cu for in vivo dual micro-PET/CT in nude mice orthotopically implanted with EPHB4-expressing U251 and EPHB4-negative U87 human GB cells. In U251 tumours, the labelled peptide co-localized with both tumour blood vessels and tumour cells; in U87 tumours, the tracer co-localized with tumour blood vessels only, not with tumour cells, indicating that the labelled EPHB4-specific peptide might be a possible tool for imaging both GB growth and neo-angiogenesis (14). A specific problem associated with ⁶⁴Cu-labelled polypeptides relates to the instability of the copper chelates and efforts are being made to develop new bi-functional chelating agents for preparing more stable ⁶⁴Cu-labelled polypeptides.

Tracers Using ¹⁸F

¹⁸F-7-Chloro-N,N,5-trimethyl-4-oxo-3(6-fluoropyridin-2-yl)-3,5-dihydro-4H-pyridazino[4,5-b]indole-1-acetamide (¹⁸F-14). As aforementioned [see ¹¹C-(R)PK11195] the 18-kDa mitochondrial TSPO is up-regulated in HGG (9) (Figure 1A). Focused structure–activity relationship studies have recently led to the development of a further fluorinated TSPO ligand with specific affinity identified by the number “14”. “14” was radiolabeled with fluorine-18 (¹⁸F-14) in fair yield and specific activity and in vivo studies on ¹⁸F-14 as a probe for molecular imaging of TSPO-expressing HGG tumours are ongoing (15). One additional TSPO ¹⁸F-ligand developed for HGG PET will be discussed later (¹⁸F-VUIIS1008).

¹⁸F-Trans-1-amino-3-fluorocyclobutanecarboxylic acid (¹⁸F-FACBC). This is a synthetic amino acid analog for PET that is currently being tested in phase II clinical trials for prostate cancer (http://aojnmb.mums.ac.ir/article_2842_398.html). The mechanisms of uptake of ¹⁸F-FACBC in C6 glioma rat cells and inflammatory cells have been investigated in comparison to those of ¹⁸F-FDG (16). Over half of ¹⁸F-FACBC uptake by glioma C6 cells was mediated by Na⁺-dependent amino acid transporters and ¹⁸F-FACBC uptake ratios of granulocytes/macrophages to tumour cells were lower than those for ¹⁸F-FDG, indicating that PET with ¹⁸F-FACBC might be advantageous in discriminating inflamed regions from tumours (16).

¹⁸F-Fluorobutyl ethacrynic amide (¹⁸F-FBuEA-GS). Lipocalin-type prostaglandin D synthase (L-PGDS) expression has been correlated with the progression of neurological disorders including some brain tumours. The imaging potency of a glutathione (GS) conjugate of the L-PGDS ligand, ¹⁸F-FBuEA-GS, for HGG was investigated in a rat model (17). ¹⁸F-FBuEA-GS was found to bind specifically to L-PGDS and the PET imaging suggested that the radiotracer accumulation in tumour lesions was mediated by the GS moiety and related to the overexpression of L-PGDS (17).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Clinical and preclinical studies with recently employed, non-routine positron-emission tomography (PET) tracers. A: Left: Co-registered and fused postcontrast T1-weighted magnetic resonance (MR) images (gray-scale) and parametric binding potential (BP_ND) images (colour spectrum) of representative cases of low-grade astrocytoma (LGA), low-grade oligodendroglioma (LGO) and glioblastoma (GBM). BP_ND is low in LGA, whereas foci with high BP_ND are found in LGO and areas of high BP_ND in GBM (arrows). Right: Co-registered postcontrast T1-weighted MR images and parametric BP_ND images in four HGGs showing little or no contrast enhancement (arrows) and high ¹¹C-(R)PK11195 binding within tumours. Colour bars denote BP_ND values. From (9) with permission. B: PET/computed tomographic (CT) imaging and biodistribution of ⁶⁴Cu-NOTA-AC133 MAB in mice bearing orthotopic U251 glioma xenografts. Nude mice bearing orthotopic xenografts of U251 glioma cells overexpressing CD133 or orthotopic xenografts of CD133–negative U251 wild-type cells received 7.5±0.8 MBq ⁶⁴Cu-NOTA-AC133 MAB via tail vein injection and PET/CT images were acquired 24 and 48 h later. Representative contrast-enhanced micro-CT and fused micro-PET/CT sections from mice bearing CD133-overexpressing U251 gliomas (left, centre) and from mice bearing U251 wild-type tumours (right) are shown. For the PET/CT images, an upper threshold corresponding to the maximum tracer uptake of the CD133-overexpressing U251 tumours was chosen. From (13) with permission.

¹⁸F-Fluoro-pivalic acid (¹⁸F-FPIA). In addition to increased glycolytic flux, tumour cells may display aberrant lipid metabolism. Pivalic acid is a short-chain, branched carboxylic acid used to increase oral bioavailability of prodrugs. After prodrug hydrolysis, pivalic acid undergoes intracellular metabolism via the fatty acid oxidation pathway. The new fluorinated probe ¹⁸F-FPIA was designed for the imaging of aberrant lipid metabolism in HGG tissues (18). Compared with ¹⁸F-FDG in an animal model of U87 tumour, ¹⁸F-FPIA had lower normal-brain uptake resulting in a superior T/N ratio and higher contrast for brain cancer imaging. Both radiotracers showed yet significant unspecific localization in inflammatory tissues (18).

¹⁸F-1-(2-Fluoro-1-[hydroxymethyl]ethoxy)methyl-2-nitroimidazole (¹⁸F-FRP170). This compound has been described as a further hypoxic cell tracer for GB (19). Patients underwent PET with ¹⁸F-FRP170 before tumour resection and the mean SUV was calculated at tumour regions showing high or low accumulation of ¹⁸F-FRP170. In those areas, intratumoural pO₂ was also measured using microelectrodes during tumour resection. The mean pO₂ was significantly lower in the areas of high uptake than in those of low uptake, suggesting that high accumulation of ¹⁸F-FRP170 might indicate viable hypoxic tissues in GB.

¹⁸F-Sodium fluoride (¹⁸F-NaF). One rare case of osseous metastases from GB diagnosed by ¹⁸F-NaF PET/CT has been reported (20). A 30-year-old man with a history of GB presented with bone pain and underwent ¹⁸F-NaF PET/CT for further evaluation. Intense uptake in two bone lesions was noted and histopathological evaluation revealed osseous metastases from GB. Hence, albeit rare, bone metastases have to be considered in patients with axial tumours and atraumatic bone pain and ¹⁸F-NaF PET/CT may be taken into consideration for their detection (20).

¹⁸F-2-(5,7-Diethyl-2-(4-(2-fluoroethoxy)phenyl)pyrazolo[1,5-a]pyrimidin-3-yl)-N,N-diethylacetamide (¹⁸F-VUIIS1008). Besides the aforementioned ¹¹C-(R)PK11195 and ¹⁸F-14, one further high-affinity TSPO PET ligand, ¹⁸F-VUIIS1008, has been evaluated in healthy mice and HGG-bearing rats. Dynamic PET data were acquired after ¹⁸F-VUIIS1008 injection, with binding reversibility and specificity evaluated by non-radioactive ligand displacement or blocking. Imaging was validated by histology and immunohistochemistry. ¹⁸FVUIIS1008 exhibited rapid uptake in TSPO-rich organs, as well as elevated T/N and binding potential in tumour tissues. Blocking of uptake and binding could be obtained by competing treatment with non-radioactive VUIIS1008 (21).

Tracers Using ⁶⁸GA

⁶⁸Ga-NOTA-15-amino-4,7,10,13-tetraoxapentadecanoic acid (PEG4)-cyclo (GDGEAyK) (⁶⁸Ga-A2B1). Integrin α2β1 may be highly expressed in HGG and is currently being explored as a prognostic biomarker for these tumours. A novel ⁶⁸Ga-labeled integrin α2β1-targeted PET tracer ⁶⁸Ga-A2B1 was designed and evaluated in preclinical models of GB. ⁶⁸Ga-A2B1 PET was performed in subcutaneous U87 tumour-bearing athymic mice and receptor targeting specificity was confirmed by competition blocking. However, due to quick renal washout and lability of peptide tracers in the blood stream, the focused ultrasound-mediated delivery method was adopted to enhance tumour uptake and retention of tracers. The average radioactivity accumulation within a tumour was quantified from the multiple region of interest volumes and was analysed in accordance with the ex vivo autoradiographic and pathological data. A significant increase in tumour uptake of ⁶⁸Ga-A2B1 was observed following focused ultrasound treatment, indicating a possible (although difficult to be clinically applied) enhancement procedure for PET imaging of HGG (22).

⁶⁸Ga-DOTA-tyr3-octreotide (⁶⁸GaDOTATOC). Somatostatin receptor subtype 2 (SSTR2) has been proposed as a potential target in HGG and its visualization and quantification could be useful for development of SSTR2-targeted therapies. Rat BT4C malignant glioma cells were injected into rat brain or subcutaneously into nude mice in order to investigate the tumour uptake of ⁶⁸GaDOTATOC, a somatostatin analogue binding to SSTR2 (23). Albeit high T/N and tumour-to-muscle ratios of ⁶⁸Ga-DOTATOC were observed by autoradiography, tumour signals detected in vivo were too low, thus compromising PET imaging and confirming the indispensability of animal investigations in this kind of studies (23). DOTATOC has been also explored in radioguided surgery (RGS) of HGG. This technique is based on selective delivery of beta⁻ radiation to the tumour by a tumour-specific emitting molecule. A study of the uptake of the beta⁻-emitting molecule ⁹⁰Y-DOTATOC and a feasibility study of the RGS technique in HGG patients have been reported (24). Uptake and background were estimated using the surrogate molecule ⁶⁸Ga-DOTATOC assuming that gallium and yttrium are chemically similar enough for the kinetics of the tracer not to differ. The T/N of ⁶⁸Ga-DOTATOC in HGG was more than 4, implying that the tracer might be selective enough for RGS (24).

⁶⁸Ga-DOTA-conjugated Z0959 affibody (⁶⁸Ga-DOTA-Z09591). Overexpression and oversignalling of the angiogenesis biomarker/tyrosine kinase receptor platelet-derived growth factor receptor beta (PDGFRβ) has been observed in multiple malignant tumours including HGG (25). Recently, the affibody molecule Z09591 labelled with ¹¹¹In has been demonstrated to be able to specifically target PDGFRβ-expressing tumours for in vivo imaging. ⁶⁸Ga-DOTA-Z09591 has shown the capacity to specifically bind to PDGFRβ-expressing U87 GB cells in vitro (26). In vivo, ⁶⁸Ga-DOTA-Z09591 demonstrated specific targeting of U87 GB xenografts in immunodeficient mice with a remarkable tumour-to-blood activity ratio of 8.0. Micro-PET imaging consistently provided high-contrast imaging of U87 GB xenografts (25).

⁶⁸Ga-S-2-(4-Isothiocyanatobenzyl)-1,4,7-triazacyclononane-1,4,7-triacetic acid-PRGD2 (⁶⁸Ga-PRGD2). The integrin αvβ3 is overexpressed in both neovasculature and glioma cells. ⁶⁸Ga-PRGD2 has been described as a new agent for non-invasive imaging of αvβ3 in patients with HGG (27). ⁶⁸Ga-PRGD2 specifically accumulated in brain tumours that were rich in αvβ3 and other neovasculature markers, but not in brain parenchyma other than the choroid plexus. ⁶⁸Ga-PRGD2 PET/CT demarcated the tumour more specifically than ¹⁸F-FDG PET/CT and the maximum SUV of ⁶⁸Ga-PRGD2 correlated with the glioma grading (27).

Tracers Using ¹²⁴I

¹²⁴I-8H9 monoclonal antibody. Convection enhanced delivery (CED) is a promising, albeit tricky drug delivery mode for treating brain tumours (28,29). Thorough measurement of the volume of distribution (V_D) of CED-infused agents in the human brain is crucial for successful treatment. ¹²⁴I may be a suitable radionuclide for in vivo measurement of volume of distribution via PET and due to the potential utility of radioimmunotherapy, ¹²⁴I-8H9 delivered via CED has been evaluated as a ‘theragnostic’ agent against pediatric tumour diffuse intrinsic pontine glioma (DIPG). In preclinical dosimetric studies, rats were subjected to CED of ¹²⁴I-8H9 to the pons with serial micro-PET performed during the subsequent 7 days. Two primates were also subjected to CED of ¹²⁴I-8H9 to the pons for safety and dosimetric analyses. The mean V_D was 0.14 cc³ in the rat and 6.2 cc³ in the primates (30) and the activity decreased 10-fold 48 h after CED in both animal models. Further studies on CED of ¹²⁴I-8H9 for DIPG might be warranted.

Tracers Using ¹³N

¹³N-Ammonia (¹³N-NH₃). A perfusion–metabolism-coupled analytical procedure with ¹³N-NH₃ and ¹⁸F-FDG PET/CT has been reported in recurrent GB (31). ¹³N-NH₃ PET/CT and ¹⁸F-FDG PET/CT have been found concordant in essentially all investigated patients. Overall sensitivity, specificity, predictive values and accuracy were >75% for both ¹³N-NH₃ PET/CT and ¹⁸F-FDG PET/CT. The performance of ¹³N-NH₃ PET/CT and ¹⁸F-FDG PET/CT were not significantly different between HGG and LGG and a positive correlation was noted between the SUV derived in the two modalities. It has thus been proposed that perfusion analysed with ¹³N-NH₃ and metabolism analysed with ¹⁸F-FDG may be coupled in HGG analysis, targeting two different but interrelated aspects of the same pathological process (31).

Prospects for Non-routine PET in Imaging of HGG

A number of issues must be considered when PET is used for imaging HGG: the variety of PET machines and tracers, the heterogeneity of the protocols used for image acquisition and the existence of semi-quiescent but highly invasive GIC (32). Specific concerns for preclinical studies of micro-PET relate to the frequent use of orthotopic tumour animal models developed by intracranial injection of established, non-GIC lines. The poor infiltrating and neovasculariing capacities of orthotopic tumours developed by established cell lines such as U87 may possibly bias radiotracer biodistribution and uptake in micro-PET studies (33). Dealing with these issues may include the harmonisation of protocols and software applications as well as the use of primary GIC lines that more closely mimic clinical tumour features (33). That said, following thorough radiopharmaceutical development, PET with specific tracers may provide important information on tumour metabolism and response to therapies as well as for determining the pharmacokinetics of novel drugs. Merely to quote one pair of examples, the pharmacokinetics of ataxia telangiectasia mutated inhibitors (ATMi) that are capable of specifically radiosensitizing GIC-driven tumours (34) or peptides inhibiting the myelocytomatosis (MYC) oncoprotein (MYCi Omomyc peptides) (35) that are promising targeting chemotherapeutics, could be conveniently studied by PET. After rational drug design by in silico modelling, labelling the basic ATMi KU60019 molecule or the basic MYCi Omomyc peptide with appropriate radionuclides (e.g. ¹⁸F) will allow investigation of the biodistribution and pharmacokinetics of these novel therapeutic agents by static and dynamic PET in the preclinical and then clinical settings (28, 34-37).