MRI: Multiparametric imaging of brain tumors

Accepted Manuscript Title: PET/MRI: Multiparametric Imaging of Brain Tumors Authors: Jiˇr´ı Ferda, Eva Ferdov´a, Ondˇrej Hes, Jan Mraˇcek, Jan Baxa PII: DOI: Reference:

S0720-048X(17)30086-4 http://dx.doi.org/doi:10.1016/j.ejrad.2017.02.034 EURR 7754

To appear in:

European Journal of Radiology

Received date: Revised date: Accepted date:

24-11-2016 19-2-2017 20-2-2017

Please cite this article as: Ferda Jiˇr´ı, Ferdov´a Eva, Hes Ondˇrej, Mraˇcek Jan, Baxa Jan.PET/MRI: Multiparametric Imaging of Brain Tumors.European Journal of Radiology http://dx.doi.org/10.1016/j.ejrad.2017.02.034 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

PET/MRI: Multiparametric Imaging of Brain Tumors PET/MRI: Multiparametric Imaging of Brain Tumors Jiří Ferda*, Eva Ferdová*, Ondřej Hes**, Jan Mraček*** , Jan Baxa* Clinic of the Imaging Methods* Sikls´Institute of Pathological Anatomy** Clinic of the Neurosurgery*** University Hospital Plzen Alej Svobody 80 304 60 Plzeň, Czech Republic

Jiří Ferda, M.D., Ph.D., Prof. Clinic of the Imaging Methods University Hospital Plzen Alej Svobody 80 304 60 Plzeň, Czech Republic [email protected]

Eva Ferdová, M.D.. Clinic of the Imaging Methods University Hospital Plzen Alej Svobody 80

2

304 60 Plzeň, Czech Republic [email protected]

Ondřej Hes, M.D., Ph.D., Prof. SIkls´Institute of Pathological Anatomy University Hospital Plzen Alej Svobody 80 304 60 Plzeň, Czech Republic [email protected]

Jan Mraček, M.D., Ph.D., Assoc. Prof. Clinic of the Neurosurgery University Hospital Plzen Alej Svobody 80 304 60 Plzeň, Czech Republic [email protected]

Jan Baxa, M.D., Ph.D., Assoc. Prof. Clinic of the Imaging Methods University Hospital Plzen Alej Svobody 80

3

304 60 Plzeň, Czech Republic [email protected]

Highlights



A combination of morphological imaging of the brain with microstructural and functional imaging provides a comprehensive overview of the properties of individual tissues. While diffuse imaging provides information about tissue cellularity, spectroscopic imaging allows us to evaluate the integrity of neurons and possible anaerobic glycolysis during tumor hypoxia, in addition to the presence of accelerated synthesis or degradation of cellular membranes; on the other hand, PET metabolic imaging is used to evaluate major metabolic pathways, determining the overall extent of the tumor (18F-FET, of differentiation (18F-FDG,

18F-FDOPA, 18F-FCH)

18F-FLT,

18F-FDOPA

and

or the degree

18F-FET).

Multi-

parameter analysis of tissue characteristics and determination of the phenotype of the tumor tissue is a natural advantage of PET/MRI scanning. The disadvantages are higher cost and limited availability in all centers with neuro-oncology surgery. PET/MRI scanning of brain tumors is one of the most promising indications since the earliest experiments with integrated PET/MRI imaging systems, and along with hybrid imaging of neurodegenerative diseases, represent a new direction in the development of neuroradiology on the path towards comprehensive imaging at the molecular level.

4

Abstract A combination of morphological imaging of the brain with microstructural and functional imaging provides a comprehensive overview of the properties of individual tissues. While diffusion weighted imaging provides information about tissue cellularity, spectroscopic imaging allows us to evaluate the integrity of neurons and possible anaerobic glycolysis during tumor hypoxia, in addition to the presence of accelerated synthesis or degradation of cellular membranes; on the other hand, PET metabolic imaging is used to evaluate major metabolic pathways, determining the overall extent of the tumor ( 18F-FET, FDOPA,

18F-FCH)

or the degree of differentiation (18F-FDG,

18F-FLT, 18F-FDOPA

and

18F-FET).

18F-

Multi-

parameter analysis of tissue characteristics and determination of the phenotype of the tumor tissue is a natural advantage of PET/MRI scanning. The disadvantages are higher cost and limited availability in all centers with neuro-oncology surgery. PET/MRI scanning of brain tumors is one of the most promising indications since the earliest experiments with integrated PET/MRI imaging systems, and along with hybrid imaging of neurodegenerative diseases, represent a new direction in the development of neuroradiology on the path towards comprehensive imaging at the molecular level.

Introduction The brain, as the most metabolically active tissue of the human body, is characterized by its variable behavior and often overlapping changes at the metabolic, structural and functional levels. Considering the brain either in structural or functional terms only is inadequate in assessing the pathophysiological chain of cause and effect. In contemporary medicine, diseases of the central nervous system are one of the most important challenges of diagnostic imaging. The guiding principle is multiparametric imaging and multifactorial analysis of the tumor phenotype [1]. A combination of morphologic, metabolic, microstructural and functional approaches to imaging is currently optimally enabled by combining positron emission tomography with magnetic resonance imaging. The population incidence of brain tumors has had a growing trend since 1970s to 1990s [2]. In the western population, the number of patients with neuroepithelial tumors is being steadily the important cause of cancer-based mortality [3]. Histological type, the level of differentiation and the possibility to perform radical removal of the tumor are the most important factors for the prognosis of patients. Therefore, there have been increasing demands on preoperative imaging of the tumor tissue itself, determination of the overall extent of the disease and assessment of the location to collect a biopsy sample for histological diagnosis. In addition to assessing the tumor itself, it is very important to visualize the relationship of the tumor process to the functional centers and white matter tracts. In patients who have already undergone surgical resection or have received a combination of chemotherapy and radiotherapy, it is important to evaluate the presence of residual tumor tissue, recognize potential changes in tissue behavior, secondary changes in the brain tissue related to the presence of radionecrosis and

5

other non-tumor reparative processes after the treatment. Magnetic resonance imaging has been the gold standard for neuro-oncology diagnosis, where functional and multiparametric imaging have their role in advanced imaging algorithms [4].

Patient preparation and data acquisition The basic requirement in the preparation of patients before PET/MRI scanning is accumulation of the radiopharmaceutical under resting basal conditions. The patient should be placed on bed resting, preferably in a darkened room in silence. The time of radiopharmaceutical accumulation varies between different substances. In addition to general measures during the accumulation of the substances, it is also important to adhere to fasting requirements in patients. This is particularly important when using

18F-FDG,

which features a significant competition with glucose. Therefore, at least four hours of fasting are required prior to its administration. For other substances, fasting has limited importance in influencing the accumulation of substances, but fasting can minimize the risks associated with the development of nausea during the examination. PET brain scanning is performed using an integrated scanner with a PET detector insert made of lutetium silicate. The acquisition itself is performed with simultaneous PET and MRI scanning in the magnetic resonance head coil with twelve receiving channels. The first step is the localization scan. The localization is followed by a T1-weighted sequence of VIBE two-point-Dixon gradient echo. The obtained in-phase, opposed-phase images, and calculated fat-images and water-images are used to create a tissue model for attenuation correction and subsequently to reconstruct the attenuation-corrected PET images. The initial sequence is followed by PET data acquisition. PET images are used to evaluate the metabolic processes in the brain tissue based on data usually acquired during the 15 to 20 minutes of continuous PET data acquisition. However, data can also be acquired dynamically on a continuous time axis from intravenous and development of the time dependency of the radiopharmaceutical accumulation monitored. Dynamic examinations require long data acquisition, typically from 20 to 40 minutes, and the acquisition starts with the beginning of the actual administration of the radiopharmaceutical. For radiopharmaceuticals with a dual distribution pattern, i.e., perfusion-dependent distribution and late tissue-specific steady-state distribution, biphasic acquisition is a possible variant. Being located in the neurocranium, the brain is a tissue without independent distinctive movements that would lead to distortion of spatial information or distortion of the space. However, precise assessment of the metabolic activity of the gray and white matters has revealed that pulsating movements of the brain tissue cause an impairment of the spatial resolution of PET scan with a "blurred" activity. These problems could be solved using segmented motion-correction reconstruction[4]. At present, the data acquisition by echoplanar BOLD (blood-oxygen-level-dependent) is used to segment PET data and to correct this type of artifact (COMPASS, Siemens Healthcare).

6

Brain MRI Standard protocol using T2-weighted sequences of fast spin-echo, T2-weighted FLAIR (fluid attenuation inversion recovery) sequences, diffuse imaging using echoplanar sequences with calculation of the map of apparent diffuse coefficient (ADC) and T1-weighted gradient echo sequences (FLASH - fast low angle shot single or MPRAGE - magnetization prepared rapid gradient echo) is used for structural imaging in PET/MRI scans of the brain. Detailed macrostructural morphological imaging still brings basic and pivotal information. Susceptibility weighted imaging (SWI) is an advantageous tool to visualize the deposits of hemoglobin degradation products, which benefits from extreme changes in T2* signals for hemosiderin and deoxyhemoglobin and can be used to detect signs of bleeding. Many studies suggest a positive correlation between the histologically established grade of glial tumor and diffusion restriction, which means a negative correlation between tumor grade and the value of the apparent diffusion coefficient (ADC) [6], the diffusion restriction also being an important feature of brain lymphomas. Diffusion-weighted imaging in the primary diagnosis significantly contributes to finding tumors of small dimensions, and increases sensitivity and specificity in the detection and differential diagnosis of brain tumors [6]. Since a PET/MRI scan of the brain is usually performed in patients for whom information on the structure and level of organization of the brain’s white matter in the surrounding of the tumor tissue may be significant, the standard diffusion weighing sequence can be replaced in the imaging protocol with multidirectional diffusion weighted imaging (MDDWI). The collected data can be used to calculate the diffusion trace imaging, apparent diffusion coefficient (ADC), fractional anisotropy (FA) and diffusion tensor with the option of tractography reconstruction. Fractional anisotropy maps are important in distinguishing an expansive type of growth, when increased organization and anisotropy are paradoxically observed in the white matter around the tumor tissue, and on the contrary, decreased fractional anisotropy is seen at the site of white matter disorganization caused by infiltrative tumor growth and/or vasogenic edema. Tissue perfusion and extracellular molecular exchange During extracellular distribution, most substances follow similar behavioral principles. Gadolinium chelates are model substances traditionally used to visualize the integrity of the blood-brain barrier. Although conventional perfusion models of brain tissue magnetic resonance imaging use the T2 effect of the gadolinium contrast agent and echoplanar sequences, gradient echo sequences with rapid acquisition time can be used to obtain a set of images in the range up to steady distribution, which can be used to better evaluate and quantify certain pharmacodynamic parameters important for evaluation of the tumor tissue. It is now possible to use T1-weighted spoiled-gradient-echo VIBE (Volume Interpolated Breath Hold Examination) sequences with the acquisition of 20 to 30 series in the first minutes from the time of intravenous contrast medium administration with continuous acquisitions one after the other, or using

7

overlapping data acquisition TWIST, TWIST-VIBE and the like. Standardized contrast agent administration using an automatic injector is essential for subsequent image analysis. Gadolinium contrast agent is administered at a dose of 0.1 mmol/kg body weight at 1.5 to 2 mL/sec and flushed with 50 mL normal saline. A contrast agent with a higher concentration (one-molar gadobutrol) or higher relaxivity (gadobenate dimeglumine) may be preferable to achieve higher changes in signal intensity over time with a steeper bolus curve. Movement of circulating fluid in the tissue depends on a number of interrelated processes [6, 7]. Primarily, these are blood supply to the tissue, which means inflow through the arterial bloodstream and flow through the capillaries, while another important parameter is the exchange of fluid between the intravascular extracellular and extravascular extracellular space (i.e., between the blood plasma and tissue fluid), depending on the permeability of the vascular wall, the volume of the extravascular extracellular space and its microstructure (stroma). The essential and simplest expression of the dynamic phenomenon is visualization of the saturation curve, either as relative values or using mathematical models, as the calculated concentration of the administered contrast agent (specification of the contrast agent is required, particularly with regard to the size of the molecule). The steepness of the concentration increase of the contrast agent indicates the state of tissue vascularization, but is also suitable for comparative examinations before and after the treatment. In addition to evaluating the shape of the curve, it may be helpful to use semi-quantitative numerical parameters or their color-coded maps for the evaluation, such as the integral of the initial area under curve (iAUC), or time to peak (TTP). With the application of pharmacokinetic models (most frequently Tofts two-compartment model), the following parameters are calculated [7]: the transfer constant from the vascular extracellular space into the extravascular extracellular space (transfer constant - Ktrans), extravascular extracellular volume fraction (Ve), the transfer constant from the extravascular extracellular space back into the vascular extracellular space (K ep), or possibly other parameters when using second generation models. Generally, malignant, highly cellular tumors show a steep increase in the enhancement curve (also including a high value of iAUC), followed by variably rapid washing of the contrast agent, and an increased Ktrans value. Less cellular tumors with stroma formation, benign tissues, and in particular fibrous tissue replacing tumors during therapy, show a gradually increasing enhancement curve and lower Ktrans values [5]. According to the mechanism of action of antitumor therapy, a good early therapeutic response can be accompanied by a paradoxical increase in Ktrans, as with diffusion restriction, due to a decrease in the extracellular space caused by intracellular edema and increased vascular permeability due to local reaction (radiotherapy) [7]. According to studies, the amount of fluid exchange expressed by Ktrans shows a positive correlation with the effect of cancer therapy. Evaluation of the perfusion and permeability parameters in the brain tissue is a challenging issue, as we need to consider the impermeability of the intact blood-brain barrier for a contrast agent, or conversely, its targeted disruption following surgery or radiation insult during teleradiotherapy.

8

Neurofunctional imaging Signal changes depending on oxygen binding to hemoglobin is used for functional brain evaluation using the BOLD (blood oxygen level dependent) echo planar imaging method. Functional magnetic resonance imaging may be a part of the protocol prior to surgical resection, or as a minimal variant, motor imaging of the upper extremities (finger tapping or thumb opposition) can be combined with a speech test such as an image naming test or verbal fluency test. Functional imaging may be a part of stereotactic planning. In addition to evaluating the structure of the white matter, MDDWI may be used for tractography reconstruction, or three-dimensional modeling of white matter tracts. Tractography, like functional MRI imaging, is a procedure that allows stereotactic planning of neurosurgical procedures.

Spectroscopic analysis of metabolites Chemical shift imaging used in spectroscopy of the brain tissues enables the evaluation of individual metabolites. Magnetic resonance spectroscopy remains the only method capable of monitoring the concentration of certain metabolites in tissues in vivo, selectively, noninvasively and nondestructively as part of multiparametric MRI scanning of glial tumors [8], often complementarily with diffusion weighted imaging [9]. Hydrogen spectroscopy is routinely used in clinical practice, while other elements with an odd nucleon number, such as

13C

or

31P,

are experimental. The ability to distinguish several metabolites in the

spectrum increases with the value of magnetic induction, since magnetic induction is directly proportional to the width of the frequency band in which the measured spectrum is spread. The magnetic induction value of 3T is an advantage in PET / MRI scanning. Information obtained from spectral analysis generally has a complementary function in the diagnosis of cancer and should be consistently correlated with other parameters, as most techniques, such as single voxel spectroscopy and multivoxel chemical shit imaging, suffer from a limited volume of evaluable tissue and the quality of the analysis may be greatly influenced by local non-homogeneity of the magnetic field in vivo. Choline (Cho) is the main metabolite of interest for neuro-oncology purposes [8, 9]. The spectrum shows a choline peak at 3.22 ppm (parts per million), which includes contributions from choline itself and from a mixture of choline compounds (phosphocholine, and glycerofosfocholin phosphatidylcholine). The concentration of choline compounds in the tissue is indicative of increased membrane and phospholipid metabolism, and high cellularity. The value of the total choline concentration is generally positively correlated with tumor grading (cellularity, proliferation, dedifferentiation) [2]. For inhomogeneous tumors, spectroscopy enables the possibility of targeted biopsy. Choline, however, is not a tumor-specific parameter, and an increase in choline concentration occurs not only in the areas of increased cell proliferation, but also cell destruction. An increased choline signal in the spectrum also occurs in demyelinating diseases. The concentration ratio of choline and other metabolites specific for the tissue is advantageously used for evaluations in the brain. One of these brain-specific metabolites is N-

9

acetylaspartate (NAA) at the position of 2.02 ppm, which is a neuronal marker, and a decrease in its concentration corresponds to neuronal damage. Another traceable metabolite is lactate (Lac) at the position of 1.33 ppm, which is a marker of anaerobic glycolysis and, together with a low ADC value, reliably detects cerebral ischemia. An increase in choline signal intensity in the spectrum indicates the presence of anaerobic glycolysis in the tumor tissue, while an increased lactate signal in the spectrum indicates the presence of hypoxic adaptation in the tumor tissue, indirectly suggesting an increasing probability of tissue resistance to chemotherapy and radiotherapy. Increased Lac/Cho ratio, together with lower values of perfusion parameters, increases the sensitivity and specificity in distinguishing post-radiation necrosis of the brain from recurrent glioblastoma [4]. This information has added importance in assessing the accumulation of radiopharmaceuticals in tissues after radiotherapy. Myoinositol at the 3.5 ppm position, as a precursor of phosphatidylinositol, the main inositol-containing phospholipid, is increased in low-grade gliomas and glial responses, whereas its decrease has been reported in glioblastoma, besides changes in the lactate portion and other less apparent spectral shifts [8 9]. Magnetic resonance spectroscopy has further potential for the detection of clinically important mutations in glial tumors. It turns out that the use of advanced sequential techniques will possibly allow spectroscopy analysis with a special focus on 2-O-glutarate (2HG) at position 2.25 ppm, and could help to distinguish the important mutations of isocitrate dehydrogenase 1, which catalyzes the oxidative decarboxylation of isocitrate to 2HG in mutant glioma cells [10, 11]. Integrated imaging with PET/MRI scanning can be used to target spectroscopy imaging using either morphological magnetic resonance imaging with optional selection of the optimum position of the voxel(s) for post-contrast or diffuse imaging. Spectroscopy, however, can also be performed subsequently, after evaluating the distribution of the radiopharmaceutical in PET scanning [12].

Energy metabolism 18

F-fluorodeoxyglucose (18F-FDG) is the main marker of aerobic energy metabolism, oxidative

glycolysis in imaging methods.

18F-FDG

is a substance analogous to glucose, the completely dominant

substrate of the energy metabolism of brain tissue. It is also the most commonly used radiopharmaceutical for PET scanning. It is a substance which readily penetrates the blood-brain barrier, and its distribution is proportional to the perfusion of the brain tissue and to the level of oxidative glycolysis. This radiopharmaceutical reaches steady state relatively early in the brain tissue, earlier than in the other tissues of the human body, therefore, examination can already be performed after the accumulation about 40 minutes following the intravenous administration. Under normal circumstances, the accumulation in healthy gray matter is high. As mentioned hereinabove, the amount of distribution is proportional to the perfusion of the tissue and the level of oxidative glycolysis, which is proportional to neuronal activity in normal brain tissue. For these reasons, it is important to initiate the examination by accumulation of

18F-

10

FDG in silence, so that it is possible to minimize the activity of the speech, visual and auditory centers. In cancer patients, this resting accumulation is as important as in other examinations, such as those for the detection of epileptogenic foci or for the diagnosis of neurodegenerative diseases. Under normal circumstances, a high level of

18F-FDG

accumulation is seen in the basal ganglia, while more or less

uniform distribution is present in the cerebral cortex. Accumulation of

18F-FDG

increases with the increase in neuronal activity that is often observed in

patients in whom preparation was not performed under resting conditions, but also in areas of the cerebral cortex, which have increased activity, especially during ictal activity of the epileptogenic foci. A pathologically increased accumulation of FDG above the level of gray matter rarely occurs in cancer, e.g., in dedifferentiated tumors. Among neuroepithelial tumors, these are anaplastic astrocytoma, anaplastic oligodendroglioma, and in particular glioblastoma multiforme [13, 14]. Other tumor diseases with utilization exceeding the level in the gray matter include primary brain lymphoma and some bulky secondary brain tumors such as melanoblastoma. In dedifferentiated tumors, the high energy needs of rapidly dividing cells play a great role in the deregulation of aerobic glycolysis. The activity of glucose transporters and the intracellular activity of hexokinase are increased, while the 6P-phosphatase activity decreases. In this way, a unidirectional flow of

18F-FDG

is activated in the cells which leads to its intracellular capture

(entrapment). Only extremely active glioblastomas have such extreme tissue activity that the accumulation in the tumor tissue is higher than in the surrounding gray matter. Conversely, well differentiated neuroepithelial tumors, in particular diffuse astrocytoma, exhibit a low level of glycolysis and therefore a low accumulation of

18F-FDG.

The reason is low growth activity of

the tissue, where most of the tumor tissue is composed of white matter axons and infrequent glial elements, with a relatively small number of tumor cells scattered between them. In addition to well differentiated glioepithelial tumors, a low level of glycolysis can also be observed in meningeal tumors and also in the brain tissue of epileptogenic foci in case of interictal resting metabolism. During the resting period, the level of oxidative glycolysis decreases in such tissues below the normal resting metabolic levels of normal gray matter. Also, the level of glycolysis in reparative processes is mostly very low, and postoperative glial reactions as well as changes accompanying post-radiation necrosis are characterized by the accumulation of

18F-FDG

at the same level as the surrounding unaffected white matter, or the level

of accumulation is slightly increased [14]. 18

F-fluoromisonidazole (18F-FMISO) is a derivative of imidazole that can be used for in vivo

imaging to detect cells in tissues suffering from hypoxia. Although brain tissue is known to exclusively use aerobic glycolysis to obtain energy, a spiral of hypoxia-induced changes develops in tumor tissues at a certain stage of high-grade tumors, when lactate begins to play a role in energy metabolism under hypoxic conditions [14]. Hypoxia stimulates the production of VGEF (Vascular Endothelial Growth Factor), which is important for the stimulation of neoangiogenesis. It is a substance that is bound in reduced form to the macromolecules produced in hypoxic cells. The radiopharmaceutical is retained in the cells in which

11

reoxidation is decreased due to reduced partial pressure of oxygen, but only in viable cells with functional nitroreductase activity and not in dead cells [15]. Although information on tissue hypoxia can also be obtained from imaging techniques by analyzing the proportion of lactate in the spectrum, the potential of detecting areas of tumor tissue with a population of cells featuring tissue hypoxia by

18F-FMISO

is

primarily allowed by visualizing the part of the tumor tissue with probable resistance to conventional therapeutic procedures of teleradiotherapy and chemotherapy [16, 17]. By combining markers of oxidative glycolysis and markers of tissue hypoxia, it could be modeled the risk of resistance to therapy [18], where magnetic resonance imaging (BOLD technique for measuring oxygen consumption in tissues, or conversely for the assessment of lactate in the spectrum) could be used as one of the methods in next future, and PET scanning as the other method (18F-FDG). Although BOLD technique use is still challenging [18], especially in the light of the BOLD echoplanar sequence based motion correction of PET data, a novel approach should be tested as a combined approach in oxygen consumption and glycolysis assessment in the tumorous tissue. Proliferation activity The proliferative activity of tumor tissue is directly proportional to the population of cells that are in the preparatory or division phase of the cell cycle, and uncontrolled cell division is the main mechanism of tumor tissue growth in glial tumors. The typical processes in cell division are replication of deoxyribonucleic acid (DNA) and synthesis of cellular membranes. 18

F-fluorothymidine (18F-FLT) is a substance that can be used to monitor DNA replication in vivo.

18F-FLT

is a substance analogous to the deoxyribonucleic acid (DNA) base thymidine, which is typical just

for DNA. It is a substance that maps the metabolic chain of DNA synthesis in the synthetic phase of the cell cycle (S-phase).

18F-FLT

accumulation in the brain tissue is dependent on two partly independent

processes, cell proliferation and disruption of the blood-brain barrier. Since thymidine kinase and the transmembrane thymidine transporter are active in the S phase of the cell cycle, intracranial cell structures are normally in the quiescent G0 phase and at the same time, the blood-brain barrier is impermeable.

18F-

FLT is therefore not accumulated in normal brain tissue. However, it is unable to penetrate an intact bloodbrain barrier only in non-conjugated state. Conjugated

18F-FLT,

which is produced by conjugation with

glucuronide in the liver, is able to penetrate brain tissue.

18F-FLT

is a marker with proven high correlation

with

proliferative

activity

determined

during

the

histopathological

examination

with

the

immunohistochemical marker of proliferation Ki-67 [19, 20]. When administering

18F-FLT,

the time window between intravenous administration of the

radiopharmaceutical and the start of data acquisition for PET should be shortened as much as possible, because it is necessary to complete the examination within 15 to 20 minutes of intravenous administration in order to capture the penetration of non-conjugated After 30 to 45 minutes,

18F-FLT

18F-FLT

through the damaged blood-brain barrier.

is already conjugated and penetrates even an intact blood-brain barrier.

12

Late examination does not increase the detection capabilities of tissues with increased proliferative activity 18F-FLT

in primarily diagnosed brain tumors [19, 21], but can influence the accumulation of

processes with gliosis. Dynamic data acquisition is also possible following administration of

in reparative

18F-FLT

which

allows us to monitor the level of permeability in brain tissue or tumor tissue [20]. However, dynamic imaging fails to provide a significant information benefit compared to dynamic post-contrast magnetic resonance imaging. 18F-FLT

The most appropriate indication for

neuroepithelial tumors, where the accumulation of indication,

18F-FLT

18F-FLT

imaging is upgrade detection in low-grade is typical for early stages of conversion. In this

is an excellent tool for selecting sites for biopsy sampling and therefore it provides a

substantial contribution to correct determination of the actual grade of gliomas. The disadvantage of

18F-

FLT is dependence on an impaired blood-brain barrier. If the latter is intact, the accumulation of FLT can be low even in tumors with incidental upgrade development in low-stage tumors. On the contrary, in significant glial reactions in late forms of glial response to a radiation insult, the accumulation of FLT may be present even in gliomatous glial reparative responses to radionecrosis. In contrast, special caution is needed when evaluating the effect of antiangiogenic therapy in high-grade gliomas following the administration of bevacizumab, as the permeability may return to normal levels and functional impairment of the blood brain barrier may resolve [22]. 18

F-fluorocholine (18F-FCH) is primarily a marker of lipid metabolism, and is used in the diagnosis

of cancer as a marker of the synthesis of cellular biphospholipid cell membranes. In PET/MRI imaging, the importance of

18F-FCH

as a radiopharmaceutical decrease, as it provides complementary information like

the choline analysis in the MRS spectrum. In glial tumors of the brain, the use of

18F-FCH

is beneficial,

because as a lipophilic substance, it penetrates the blood-brain barrier and accumulates in the tumor tissue depending on the rate of synthesis of cell membranes [14, 23].

18F-FCH

is a partly equivalent

marker of cell proliferation, even though it indicates the synthesis of cell membranes, not only the cell wall. Choline can also be used to visualize low-grade components of intracranial tumors, where an increased accumulation of FCH in the tissue of low-stage glioma is correlated with microvascular density in the tumor, when an impairment of the blood-brain barrier is not necessarily required [14]. Protein synthesis Protein synthesis is deregulated in brain tumors. The expression of the genes encoding large amino acid transporters, such as LAT1, is increased compared to normal brain tissue. Increased protein synthesis is proportional to the increase in proliferative activity, and serves as a marker of accelerated cell growth. In the diagnosis of cancer, PET scanning is used preferably with three radiopharmaceuticals, analogues of phenylalanine, methionine, and tyrosine: 18

fluorodihydroxyphenylalanine ( F-FDOPA),

11

18

F-fluoroethyltyrosine (18F-FET),

11

C-methionine ( C-MET). At present,

18F-FET

18

F-

seems to

be an optimal radiopharmaceutical, as it has a low affinity for the dopaminergic system, unlike 18F-FDOPA,

13

and can be used in centers that lack the possibility of producing radiopharmaceuticals in cyclotrons at the place of use. The advantages of FET are a significant correlation of the accumulation rate of cell density, proliferative activity and microvascular density. [22]. The use of

18F-FET

18F-FET

with

imaging for

navigating complete resection of tumor tissue significantly increases the overall survival of patients, which is an important factor encouraging the use of of

18F-FET

18F-FET

in the diagnosis of brain tumors. Similarly, the use

PET imaging for radiotherapy planning improves the accuracy of targeted irradiation during

teleradiotherapy. An important issue is differentiation of low-grade and high-grade components of glial tumors, as low-grade and high-grade tumors have significantly different kinetics. The low-grade component typically has a logarithmic shape of the accumulation vs. time curve, while the high-grade component has a shape characteristic for hypervascularized tumors with the peak accumulation level early after the administration of the radiopharmaceutical, and gradual decrease in the accumulation over time, due to the wash-out effect. This wash-out is analogous to the wash-out of contrast agents or other substances with primary extracellular distribution. Monitoring the progress of

18F-FET

accumulation in the

tissue is possible by means of dynamic imaging for a period of 40 minutes, starting at the time of administration of the radiopharmaceutical, or we can use two-phase imaging with early acquisition at 5 to 15 minutes after the administration of

18F-FET

and late acquisition at 25 to 40 minutes [24, 25, 26, 27].

Markers of protein synthesis, as substances allowing us to visualize the low-grade component of brain tumors, can be used to visualize the total volume of tumor tissue in brain glial tumors, or used for the detection of high-grade components in both primary tumors and recurrent tumors after comprehensive therapy [28], and even in tumors with heterogeneous structure with high-grade and low-grade components. An increased accumulation of

18F-FET

can be observed in high-grade tumors, even past the

borders of tissue saturated with gadolinium contrast agent to visualize the known phenomenon, the presence of glioblastoma tumor cells far beyond the limit of macroscopic tumor detection. A similar mechanism to that of and

11C-MET

while in

18F-FET

is also used to image glial brain tumors by

18F-FDOPA

[29, 30, 31, 32, 33]. Methionine has a very similar mechanism of metabolism as

18F-FDOPA

18F-FET,

imaging, FDOPA is converted to dopamine by the action of dihydroxyphenylalanine

decarboxylase, or to 3-O-methyl-6-fluoro-L-dihydroxyphenylalanine (3-OMFD) by the action of catechol-Omethyl transferase. The use of

18F-FDOPA

imaging of glial tumors, in particular high-grade tumors, is

associated with a problem, a high degree of accumulation in the dopaminergic system in the brain/basal ganglia, but also in the putamen or substantia nigra, while in the cortical and sub-cortical areas, no superposition of activity with normal tissue is present. An increase in

18F-FDOPA

and

11C-MET

turnover in

the tissue also serves as a marker of increased protein synthesis, because increased transportation of amino acids is always the main component of the metabolic chain in brain tumors, and their increased turnover may also serve as a prognostic sign or guide to detecting local recurrence or persistence in the viable tumor cell population [29, 30, 34]. Synergistic PET/MRI imaging

14

The key question is, what is the optimal indication of PET/MRI imaging in brain tumors? The related question is what radiopharmaceutical should be used. Besides the specific characteristics of each radiopharmaceutical, such as the unique use of

18F-FLT

as a double marker indicating an impaired blood-

brain barrier and increased proliferation, a number of characteristics are highly complementary. On the one hand, the use of

18F-FDG

in PET imaging is significantly disqualified by the high metabolic

background of the brain tissue, but on the other hand, it allows us to correlate the positions of normal and pathological tissues when combining precise merger with MRI imaging. Since

18F-FDG

shows a very

similar specificity and sensitivity in differentiating low-grade and high-grade glial tumors, like other radiopharmaceuticals, its great advantage is high availability and relatively low cost compared to other radiopharmaceuticals. A substantial benefit of high-grade tumors, and therefore

18F-FDG

18F-FDG

is high specificity in differentiating the presence of

can be used very effectively in distinguishing post-radiation

necrosis from local recurrence of high-grade tumors. Some radiopharmaceuticals with high reliability in newly diagnosed cancers (such as

18F-FLT,

but also

18F-FET

or

18F-FDOPA)

may paradoxically have

problems in distinguishing slow kinetics in the areas of reactive gliosis, when used for imaging of long accumulation times. Therefore, the use of early imaging may be advisable in recurrent high-grade tumors (or combined with late imaging in the case of

18F-FET

and

18F-DOPA)

and distribution of the

radiopharmaceutical should be correlated to the presence of diffusion restriction in ADC maps [35, 36], or spectroscopy should be targeted according to the accumulation of the radiopharmaceutical [7]. The indication of

18F-FLT

accumulation of

seems to be optimal for the detection of high-grade tumor tissue. The high

18F-FLT

is very closely associated with the level of Ki-67 positivity, which is the

immunochemical marker of proliferative activity. It is highly advantageous to use

18F-FLT-PET/MRI

in

cases where the presence of a high-grade component is to be detected, in diseases with suspected tumor up-grade, or in cases where stereotactic biopsy of the tumor tissue is considered. The use of

18F-FLT-

PET/MRI for biopsy navigation reduces the probability of misdiagnosing low-grade glioma when taking a sample from the low-grade component [19, 20]. Given the relatively long period of data acquisition for PET/MRI imaging, false-positive findings, if any, may rarely be based on a concordant increase in the accumulation of FLT in low-grade tumors with an impaired blood-brain barrier, for example, some ependymomas. The accumulation of the concentration of

18F-FLT

18F-FLT

in these cases may result from a passive gradual increase in

in the tissue without any active accumulation in the proliferating cells. For

similar reasons, it may be difficult to detect the presence of persistent tumor tissue in patients after resection or teleradiotherapy of high-grade tumors [37]. Escalated glial reaction and the different kinetics of

18F-FLT

and aminoacid analogues in the absence of the blood-brain barrier may be the cause of false

positive findings of radiopharmaceutical accumulation in radiation necrosis or gliomatous response. At present, a group of substances derived from amino acids seems to be very promising for glial tumors, according to the frequency of literature references. The use of

18F-FET

and

18F-FDOPA

the primary detection of glial tumors, amino acid analogues, compared to the use of

may be similar. In

18F-FDG

and 18F-FLT,

can be used to visualize whole tumors, including their low-grade components. In this case, the

15

accumulation of radiopharmaceuticals for at least 20 minutes is predominantly used, and this technique allows us to navigate for resection of low-grade tumors by volumetry [37] and also to detect the residual tissues of low-grade gliomas after resection [33]. Aminolevulinic acid hydrochloride (Gliolan) can be used for resection of high-grade tissues, including recurrent tumors, and both

18F-FLT

imaging and amino acid

analogues seem to be effective due to the high degree of precision. The extent of tissue with fluorescence correlates well with areas of high accumulation of the radiopharmaceutical. The most important information obtained from imaging of neuroepithelial brain tumors is the real extent of tumor infiltration in the brain tissue, the degree of tumor differentiation, the relationship of the tumor tissue to functionally important structures of the brain, and last but not least, prediction of the disease prognosis. The following information is important in patients who have already undergone treatment for brain cancer: persistence of viable tumor tissue, signs of recurrence, and information on the presence of non-tumor reparative changes and/or post-radiation necrosis. Combined PET imaging and multiparametric magnetic resonance imaging is advantageous as it allows us to complete the information obtained from pharmacodynamic analysis, which fully replaces dynamic PET imaging, allows us to detect perfusion changes in tissue with impaired permeability, evaluate changes in the size of the extracellular volume, even in combination with perfusion compartmental analysis and assessment of diffuse imaging. Evaluation of the choline and lactate portion in the spectrum provides additional information that can be used to distinguish between post-radiation necrosis and recurrent high-grade glioma. The expansive or infiltrative character of tumor tissue growth can be evaluated by assessing the fractional anisotropy. Combined imaging of functional centers, reconstruction of white matter tracts, and visualization of the volume of tumor tissue accumulating the radiopharmaceutical are used for planning resection procedures.

Conclusion A combination of morphological imaging of the brain with microstructural and functional imaging provides a comprehensive overview of the properties of individual tissues. While diffuse imaging provides information about tissue cellularity, spectroscopic imaging allows us to evaluate the integrity of neurons and possible anaerobic glycolysis during tumor hypoxia, in addition to the presence of accelerated synthesis or degradation of cellular membranes; on the other hand, PET metabolic imaging is used to evaluate major metabolic pathways, determining the overall extent of the tumor ( 18F-FET, 18F-FDOPA, 18FFCH) or the degree of differentiation (18F-FDG,

18F-FLT, 18F-FDOPA

and

18F-FET).

Multi-parameter

analysis of tissue characteristics and determination of the phenotype of the tumor tissue is a natural advantage of PET/MRI scanning. The disadvantages are higher cost and limited availability in all centers with neuro-oncology surgery. PET/MRI scanning of brain tumors is one of the most promising indications since the earliest experiments with integrated PET/MRI imaging systems, and along with hybrid imaging of

16

neurodegenerative diseases, represent a new direction in the development of neuroradiology on the path towards comprehensive imaging at the molecular level.

To the Editorial Board, European Journal of Radiology We wish to confirm there has been no significant financial support for this work that could have influenced its outcome. All authors have read and approved submission of this manuscript. This work has not been published and is not being considered for publication elsewhere. Conflict of interest: Dr. Ferda and Dr. Baxa are consultants for and receives research support from Siemens. The other author have no conflict of interest to disclose.

Acknowledgement: supported by project Conceptual development of the research institution of the Czech Ministery of Health No. 00669806 – FN Plzeň and by Program of the development of the Charles´ University (project P36).

17

References

[1] Tempany CM, Jayender J, Kapur T, et al. Multimodal imaging for improved diagnosis and treatment of cancers. Cancer. 121(6) (2015) 817-27.

[2] Deltour I, Johansen C, Auvinen A. et al. Time trends in brain tumor incidence in Denmark, Finland, Norway and Sweden, 1974-2003. J Natl. Cancer Inst 101(24) (2009) 1721-1721 [3] Ostrom QT, Gittleman H, Fulop J, et al. CBTRUS Statistical Report: Primary brain and central nervous system diagnosed in the United States in 2008-2012. Neuro-onkology. 17(Suppl. 4) (2015) iv1-iv62. [3] Kao HW, Chiang SW, Chung HW, Tsai FY, Chen CY. Advanced MR imaging of gliomas: an update. Biomed Res Int. (2013) 2013:970586. doi: 10.1155/2013/970586. Epub 2013 Jun 4. Review. PubMed PMID: 23862163;

[4] Catana C. Motion correction options in PET/MRI. Semin Nucl Med. 45(3) (2015) 212-223.

[5] Svolos P, Kousi E, Kapsalaki E, et al. The role of diffusion and perfusion weighted imaging in the differential diagnosis of cerebral tumors: a review and future perspectives. Cancer Imaging. 29 (2014) 1420

[6] Wang S, Kim S, Chawla S, et al. Differentiation between glioblastomas, solitary brain metastases, and primary cerebral lymphomas using diffusion tensor and dynamic susceptibility contrast-enhanced MR imaging. AJNR Am J Neuroradiol. 32(3) (2011) 507-14.

[7] Tofts PS. Modeling tracer kinetics in dynamic Gd-DTPA MR imaging. J Magn Reson Imaging. 7(1) (1997) 91-101.

[8] Yao NW, Chang C, Lin HT, Yen CT, Chen JY. Functional assessment of glioma pathogenesis by in vivo multi-parametric magnetic resonance imaging and in vitro analyses. Sci Rep.20 (2016) 6:26050. doi: 10.1038/srep26050. PubMed PMID: 27198662

18

[9]. Gadda D, Mazzoni LN, Pasquini L, Busoni S, Simonelli P, Giordano GP. Relationship between Apparent Diffusion Coefficients and MR Spectroscopy Findings in High-Grade Gliomas. J Neuroimaging. (2016) Apr 15. doi: 10.1111/jon.12350. PubMed PMID: 27079293.

[10] Gottschalk M, Troprès I, Lamalle L, Grand S, Le Bas JF, Segebarth C. Refined modelling of the short-T2 signal component and ensuing detection of glutamate and glutamine in short-TE, localised, (1) H MR spectra of human glioma measured at 3 T. NMR Biomed. 29(7) (2016) 943-951.

[11] Nagashima H, Tanaka K, Sasayama T, Irino Y, Sato N, Takeuchi Y, Kyotani K, Mukasa A, Mizukawa K, Sakata J, Yamamoto Y, Hosoda K, Itoh T, Sasaki R, Kohmura E. Diagnostic value of glutamate with 2hydroxyglutarate in magnetic resonance spectroscopy for IDH1 mutant glioma. Neuro Oncol. 18(11) (2016) 1559-1568.

[12]. Kim ES, Satter M, Reed M, Fadell R, Kardan A. A novel, integrated PET-guided MRS technique resulting in more accurate initial diagnosis of high-grade glioma. Neuroradiol J. 29(3) (2016)193-197.

[13] Dankbaar JW, Snijders TJ, Robe PA, Seute T, Eppinga W, Hendrikse J, De Keizer B. The use of (18)F-FDG PET to differentiate progressive disease from treatment induced necrosis in high grade glioma. J Neurooncol. 125(1) (2015)167-175.

[14] Lam WW, Ng DC, Wong WY, Ong SC, Yu SW, See SJ. Promising role of [18F] fluorocholine PET/CT vs [18F] fluorodeoxyglucose PET/CT in primary brain tumors-early experience. Clin Neurol Neurosurg. 13(2) (2011)156-161.

[15] Wang H, Zhang Y, Yu W, Zhao X, Xue Y, Xu H. Radiosensitizing effect of irisquinone on glioma through the downregulation of HIF-1α evaluated by 18F-FDG and 18F-FMISO PET/CT. Nucl Med Commun. 37(7) (2016)705-714. [16] Rockne RC, Trister AD, Jacobs J, Hawkins-Daarud AJ, Neal ML, Hendrickson K, Mrugala MM, Rockhill JK, Kinahan P, Krohn KA, Swanson KR. A patient-specific computational model of hypoxia-

19

modulated radiation resistance in glioblastoma using 18F-FMISO-PET. J R Soc Interface.

6 (2015)

12(103). pii: 20141174. doi: 10.1098/rsif.2014.1174. PubMed PMID: 25540239 [17]

Bell C, Dowson N, Fay M, Thomas P, Puttick S, Gal Y, Rose S. Hypoxia imaging in gliomas with

18F-fluoromisonidazole PET: toward clinical translation. Semin Nucl Med.45(2) (2015) 136-150.

[18] Valable S, Corroyer-Dulmont A, Chakhoyan A, Durand L, Toutain J, Divoux D, Barré L, MacKenzie ET, Petit E, Bernaudin M, Touzani O, Barbier EL. Imaging of brain oxygenation with magnetic resonance imaging: A validation with positron emission tomography in the healthy and tumoural brain. J Cereb Blood Flow Metab. ( 2016) Oct 4. pii: 0271678X16671965. PubMed PMID: 27702880.

[19] Ferdová E, Ferda J, Baxa J, Tupý R, Mraček J, Topolčan O, Hes O. Assessment of grading in newlydiagnosed glioma using 18F-fluorothymidine PET/CT. Anticancer Res. 35(2) (2015) 955-959. PubMed PMID: 25667480. [20] Collet S, Valable S, Constans JM, Lechapt-Zalcman E, Roussel S, Delcroix N, Abbas A, Ibazizene M, Bernaudin M, Barré L, Derlon JM, Guillamo JS. [(18)F]-fluoro-L-thymidine PET and advanced MRI for preoperative grading of gliomas. Neuroimage Clin. 29 (2015) 448-454.

[21] Zhao F, Li M, Wang Z, Fu Z, Cui Y, Chen Z, Yu J. (18)F-Fluorothymidine PET-CT for resected malignant gliomas before radiotherapy: tumor extent according to proliferative activity compared with MRI. PLoS One. 10(3) (2015) e0118769. doi: 10.1371/journal.pone.0118769. PubMed PMID: 25738617.

[22] Nowosielski M, DiFranco MD, Putzer D, Seiz M, Recheis W, Jacobs AH, Stockhammer G, Hutterer M. An intra-individual comparison of MRI, [18F]-FET and [18F]-FLT PET in patients with high-grade gliomas. PLoS One. 9(4) (2014) e95830. doi: 10.1371/journal.pone.0095830. PubMed PMID: 24759867

[23] Gómez-Río M, Testart Dardel N, Santiago Chinchilla A, Rodríguez-Fernández A, Olivares Granados G, Luque Caro R, Zurita Herrera M, Chamorro Santos CE, Lardelli-Claret P, Llamas-Elvira JM. ¹⁸FFluorocholine PET/CT as a complementary tool in the follow-up of low-grade glioma: diagnostic accuracy and clinical utility. Eur J Nucl Med Mol Imaging. 42(6) (2015)886-895.

20

[24] Jena A, Taneja S, Gambhir A, Mishra AK, Dʼsouza MM, Verma SM, Hazari PP, Negi P, Jhadav GK, Sogani SK. Glioma Recurrence Versus Radiation Necrosis: Single-Session Multiparametric Approach Using Simultaneous O-(2-18F-Fluoroethyl)-L-Tyrosine PET/MRI. Clin Nucl Med. 2016 May;41(5 (2016) e228-222

[25] Unterrainer M, Schweisthal F, Suchorska B, Wenter V, Schmid-Tannwald C, Fendler WP, Schüller U, Bartenstein P, Tonn JC, Albert NL. Serial 18F-FET PET Imaging of Primarily 18F-FET-Negative Glioma: Does It Make Sense? J Nucl Med. 57(8) (2016) 1177-1182. doi: 10.2967/jnumed.115.171033. PubMed PMID: 27033893.

[26] Albert NL, Winkelmann I, Suchorska B, Wenter V, Schmid-Tannwald C, Mille E, Todica A, Brendel M, Tonn JC, Bartenstein P, la Fougère C. Early static (18)F-FET-PET scans have a higher accuracy for glioma grading than the standard 20-40 min scans. Eur J Nucl Med Mol Imaging. 43(6) (2016) 1105-1114.

[27] Bette S, Peschke P, Kaesmacher J, Delbridge C, Pyka T, Schmidt-Graf F, Zimmer C, Meyer B, Ringel F, Gempt J. Static FET-PET and MR Imaging in Anaplastic Gliomas (WHO III). World Neurosurg. 91 (2016) 524-531.e1. doi: 10.1016/j.wneu.2016.02.094. PubMed PMID: 26947726.

[28] Moller S, Law I, Munck Af Rosenschold P, Costa J, Poulsen HS, Engelholm SA, Engelholm S. Prognostic value of (18)F-FET PET imaging in re-irradiation of high-grade glioma: Results of a phase I clinical trial. Radiother Oncol. 121(1) (2016)132-137.

[29] Sharma R, D'Souza M, Jaimini A, Hazari PP, Saw S, Pandey S, Singh D, Solanki Y, Kumar N, Mishra AK, Mondal A. A comparison study of (11)C-methionine and (18)F-fluorodeoxyglucose positron emission tomography-computed tomography scans in evaluation of patients with recurrent brain tumors. Indian J Nucl Med. 31(2) (2016) 93-102.

21

[30] Cicuendez M, Lorenzo-Bosquet C, Cuberas-Borrós G, Martinez-Ricarte F, Cordero E, MartinezSaez E, Castell-Conesa J, Sahuquillo J. Role of [(11)C] methionine positron emission tomography in the diagnosis and prediction of survival in brain tumours. Clin Neurol Neurosurg. 139 (2015) 328-333.

[31]

Yoo MY, Paeng JC, Cheon GJ, Lee DS, Chung JK, Kim EE, Kang KW. Prognostic Value of

Metabolic Tumor Volume on (11)C-Methionine PET in Predicting Progression-Free Survival in High-Grade Glioma. Nucl Med Mol Imaging. 49(4) (2015)291-297.

[32] Bell C, Dowson N, Puttick S, Gal Y, Thomas P, Fay M, Smith J, Rose S. Increasing feasibility and utility of (18)F-FDOPA PET for the management of glioma. Nucl Med Biol. 42(10) (2015) 788-795

[33] Villani V, Carapella CM, Chiaravalloti A, Terrenato I, Piludu F, Vidiri A, Schillaci O, Floris R, Marzi S, Fabi A, Pace A. The Role of PET [18F]FDOPA in Evaluating Low-grade Glioma. Anticancer Res. 35(9) (2015) 5117-5122.

[34] Karunanithi S, Sharma P, Kumar A, Gupta DK, Khangembam BC, Ballal S, Kumar R, Kumar R, Bal C. Can (18)F-FDOPA PET/CT predict survival in patients with suspected recurrent glioma? A prospective study. Eur J Radiol. 83(1) (2014) 219-225.

[35] Karavaeva E, Harris RJ, Leu K, Shabihkhani M, Yong WH, Pope WB, Lai A, Nghiemphu PL, Liau LM, Chen W, Czernin J, Cloughesy TF, Ellingson BM. Relationship Between [18F]FDOPA PET Uptake, Apparent Diffusion Coefficient (ADC), and Proliferation Rate in Recurrent Malignant Gliomas. Mol Imaging Biol. 17(3) (2015) 434-442.

[36]. Wardak M, Schiepers C, Cloughesy TF, Dahlbom M, Phelps ME, Huang SC. ¹⁸F-FLT    and ¹⁸FFDOPA PET kinetics in recurrent brain tumors. Eur J Nucl Med Mol Imaging. 41(6) (2014) 1199-209.

[37] Kobayashi K, Hirata K, Yamaguchi S, Manabe O, Terasaka S, Kobayashi H, Shiga T, Hattori N, Tanaka S, Kuge Y, Tamaki N. Prognostic value of volume-based measurements on (11)C-methionine PET in glioma patients. Eur J Nucl Med Mol Imaging. 42(7) (2015) 1071-1080.

22

Figure captions:

Figure 1: 18F-FDG-PET/MRI

of the glioblastoma in right temporal lobe, multiparametric pre-surgical examination;

tumor contains central necrosis, the high accumulation of the tumorous tissue, but the case shows the weakness of

18F-FDG

18F-FDG

is present in viable well perfused

– an insufficient delineation of the

tumorous tissue from the highly active gray matter structures. A – pharmacokinetic analysis - map of the Ktrans; B - pharmacokinetic analysis - map of the iAUC; C - 18F-FDG-PET/MRI fusion; D –functional image of BOLD (verbal fluency) showing the Brocca center within typical localization in prefrontal cortex on the left side (note the elevated 18F-FDG accumulation within center) ; E - fusion of the tractography displays the compressed intact white matter bundles between basal ganglia and tumorous tissue; F- T2 weighted FLAIR image.

Figure 2: Continued images of the case on Figure 1. A – N-acetylaspartate map of the chemical shift imaging; B – choline map of the chemical shift imaging; C – MDDWI, map of the fractional anisotropy shoving the impaired organization of the white matter in temporal lobe and the compression of the external capsule with increase fraction anisotropy; D – non-enhanced T1 FLASH image with the deposits of the

23

methemoglobin as the sign of the hemorrhage inside tumorous tissue; E – MDDWI – ADC map with restricted diffusion as a result of increased cellular density; F – MDDWI – trace image

Figure 3: Continued images of the case on Figures 1 and 2. Spectrum of the metabolites within the vowel placed in temporal lobe infiltrated by the tumorous tissue. Elevated lactate peak is the important sign of the hypoxic adaptation of the tumorous tissue.

24

Figure 4: 18F-FDG-PET/MRI

of the multifocal turnover of the diffuse astrocytoma into anaplastic astrocytoma in right

frontal lobe. Despite of focal upgrade with disrupted blood-brain barrier, low-grade component exhibits low glycolytic activity. Elevated accumulation of

18F-FDG

is localized only inside most active tissue. Motion

triggered image reconstruction using technique COMPASS (Siemens Healthcare, Erlangen, Germany) improved the spatial resolution (A,B); the conventional reconstruction (C, D) is less noisy, the accumulation is more homogenous, COMPASS reconstruction showed better the metabolic heterogeneity of tumorous tissue and enables to detect also the increased metabolic activity in the superficial enhancing tumorous tissue; E – T2 FLAIR image, F – TSE T2 image, G – enhanced MPRAGE T1; H – MDDWI – ADC map showing decreased diffusivity depending on the increased cellular density.

25

Figure 5: Continued images of the case on Figure 4. Maps of chemical shift imaging of N-acetylaspartate (A), choline (B) and lactate (C) supplemented with the spectrum enables the analysis of the myelinization breakdown (depletion of NAA), increased cellular membrane building (increased choline content) and no hypoxic adaptation (no elevated lactate).

26

Figure 6: 18F-FLT-PET/MRI

of the focal upgrade of the diffuse astrocytoma into anaplastic astrocytoma, early image

of the increased 18F-FLT distribution (C) has the same localization like the blood-brain barrier disruption on Ktrans image (A), additional maps pharmacoanalysis showing iAUC (B), Ve (D) and Kep (F). Time elapsed contrast agent distribution within the upgrade is typical with early slope and wash out, compared with the contralateral normal appeared white matter.

27

Figure 7: 18F-FLT-PET/MRI

of the diffuse astrocytoma with cystic formation in left temporal lobe. A - functional

BOLD map of the verbal fluency displaying the motor speech center on the right side on the gadolinium enhanced MPRAGE T1; B – T2 FLAIR image, C - TSE T2 image; D - no increased 18F-FLT distribution within tumorous tissue showing the low proliferation activity; E – MDDWI – map of fraction anisotropy documenting the space occupying character of the lesion compresing the surrounded white matter with consequent increased fraction anisotropy; F – MDDWI – ADC map without any sign of the diffusion restriction.

28

Figure 8: 18F-FET-PET/MRI

of the diffuse astrocytoma with cystic formation in left temporal lobe, the same patient

as on Figure 7, eighteen months later, recurrence after partial resection, re-resection confirmed the same tumor as in previous resection – diffuse astrocytoma. A - 18F-FET image with accumulation within the tumorous tissue, even if it is low-grade astrocytoma; B – T2 FLAIR image, C - MDDWI trace image; D 18F-FET-PET/MRI;

E – non-enhanced FLASH T1 image within the rest of methemoglobin signal after

previous resection; F – enhanced FLASH T1 without any enhancement within tumorous tissue.

29

Figure 9: 18F-FET-PET/MRI

of the diffuse astrocytoma with transformation into anaplastic astrocytoma. Upper row

18F-FET-PET/MRI

images with accumulation within the tumorous tissue, early phase data acquisition

between 5 – 20th minute after radiopharmaceutical application; middle row -

18F-FET-PET/MRI

with rising

accumulation within the low-grade tumorous tissue, later phase data acquisition between 30 – 45th minute after radiopharmaceutical application, lower row - MPRAGE T1 image with the application of gadolinium contrast agent showing the blood-brain barrier disruption

30

Figure 10: 18F-FET-PET/MRI

of the glioblastoma in right occipital lobe. A - 18F-FET image with the rapid accumulation

within the tumorous tissue in early phase, data acquisition between 5 – 20th minute after radiopharmaceutical application; B -

18F-FET-PET/MRI

with decreasing accumulation within the high-

grade tumorous tissue, but showing the spread of the tumor into the optic radiation, later phase data acquisition between 30 – 45th minute after radiopharmaceutical application; C – T2 FLAIR; D –TSE T2; E - FLASH T1 after application of the gadolinium contrast agent; F - MDDWI – ADC map with restriction of the diffusion within tumorous tissue.

31