| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Basic Science Investigations |
1 Nuclear Medicine Clinic, Technical University of Munich, Munich, Germany
2 Radiopharmacy, PET Center, University of Tuebingen, Tuebingen, Germany
3 Department of Nuclear Medicine, Institute H. San Raffaele, Milan, Italy
4 Cross Cancer Institute and University of Alberta, Edmonton, Alberta, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Key Words: tumor hypoxia • 18F-fluoroazomycin arabinoside • PET tracers • mouse tumor model
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Over the past 4 decades, a variety of methods to measure the hypoxic fraction in tissues have been proposed. However, none of the experimental in vivo methodologies (e.g., oxygen electrode systems) are currently used routinely in human malignancies (10), mainly because they are invasive, limited to readily accessible tumor sites such as cervix and head and neck cancer, and technically demanding. In addition, the exact localization of the probe’s tip within the tumor volume is difficult to determine. Therefore, oxygen-electrode systems are unsuitable to determine the tumor’s oxygenation distribution on a truly regional basis, which is a necessary precondition for individually adapted therapeutic approaches. Noninvasive identification and quantification of regional tumor tissue hypoxia would, therefore, be most desirable for effective treatment selection, individual treatment planning, and treatment monitoring in radiooncology.
The use of radiolabeled 2-nitroimidazoles for hypoxia imaging was proposed in 1979 by Chapman (11). Under hypoxic conditions, nitroimidazole compounds are thought to undergo reductive metabolism, forming highly reactive intermediates that consequently bind to macromolecular cellular components. Thus, these compounds tend to accumulate in sites of hypoxia and can be used for imaging purposes with both SPECT (12) and PET (13,14).
In general, a large variety of studies have been performed for evaluating different substrates with nitro- (NO2) groups. Extensive reviews of this topic are available (14,15). The nitroimidazole compound 18F-fluoromisonidazole (18F-FMISO) was the first nitroimidazole compound developed for PET and is currently being used in several cancer centers worldwide (14–19). The hypoxia-specific uptake of this radiopharmaceutical has been proven in a porcine liver occlusion model (20,21). However, its relatively slow clearance from blood and nontarget tissues results in low target-to-background ratios (T/Bk ratios) of 18F-FMISO. Other PET labeled nitroimidazole compounds with presumed hypoxia-specific retention have been reported in recent years, with some—such as 18F-fluoroerythronitroimidazole (22), 124I-iodoazomycin galactoside (23), and 18F-fluoroetanidazole (24)—showing potential in detecting tumor tissue hypoxia.
Previously, radioiodinated 1-(5-iodo-5-deoxy-
-D-arabinofuranosyl)-2-nitroimidazole (iodoazomycin arabinoside [IAZA]) was synthesized and successfully used for tumor tissue visualization (25). Compared with misonidazole, 125I-IAZA displayed rapid clearance from the blood, yielding to more favorable T/Bk ratios (26). Moreover, the arabinosyl-N1--glycosidic bond displayed enhanced in vivo stability against enzymatic cleavage. To allow tumor imaging with better spatial resolution, 18F-fluoroazomycin arabinoside (18F-FAZA) (Fig. 1) was recently synthesized and studied in various tumor-bearing mice using PET and autoradiography. In the present study, 18F-FMISO and 18F-FAZA were investigated for their suitability in detecting tumor tissue hypoxia and, in particular, to confirm the hypoxia-specific uptake mechanism for 18F-FAZA.
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
-D-arabinofuranosyl)-2-nitroimidazole in 1 mL of dimethyl sulfoxide were reacted with a mixture of azeotropically dried 18F-fluoride, 15 mg of Kryptofix 2.2.2. (Merck), and 3.5 mg of K2CO3 at 100°C for 5 min. After hydrolysis with 1 mL of 0.1N NaOH (2 min, 30°C), 0.5 mL of 0.4N NaH2PO4 was added for neutralization. The product was purified first using an alumina cartridge (Alu N light; Waters) and then high-performance liquid chromatography (HPLC) (Phenomenex Nucleosil C18; ethanol/NaH2PO4 buffer [5:95]; flow, 4 mL/min; detection ultraviolet [320 nm], NaI(Tl)). The product was obtained in radiochemical yields of 20.7% ± 3.5% (n = 12) at end of the beam. After irradiations of 1 h at 35 µA, 9.1 ± 1.4 GBq (n = 8) 18F-FAZA were isolated.
In Vivo Experiments
All animal experiments were conducted in compliance with the guidelines for the care and use of research animals established by the Animal Research Committee at the Technische Universität München. For biodistribution studies, 13 female BALB/c mice (20–25 g) were inoculated subcutaneously at the back of the thorax with a suspension of EMT6 cells (2.5 x 105 cells in 0.1 mL), a murine mammary carcinoma cell line (29), which reached the desired size of approximately 5- to 10-mm diameter in 5–8 d. Similarly, Swiss nude mice (20–23 g) were inoculated subcutaneously with a suspension of AR42J cells (30), a pancreatic acinar cell tumor, or A431 cells, a human vulva carcinoma cell line, which reached the desired tumor size of approximately 5- to 10-mm diameter after 14–21 d.
For PET studies, 13 female Swiss nude mice (20–23 g) were injected intravenously with 18F-FAZA tracer while sedated with ether. Two hours after tracer injection, PET studies were performed on anesthetized animals.
For autoradiography studies, 8 female Swiss nude mice (20–23 g) received subcutaneous xenograft implants using a tumor derived from a suspension of EMT6 cells (2.5 x 105 cells in 0.1 mL), obtaining 17 tumors of approximately 5- to 10-mm diameter at the thorax. Also, in 2 animals bearing AR42J tumors, whole-body 18F-FAZA autoradiography was performed.
Tumor Tissue Oxygenation Electrode Measurements
The presence of tumor tissue hypoxia was verified by direct tissue partial pressure of oxygen (tissue PO2) electrode measurements in 5 AR42J and 5 EMT6 tumors with their respective reference tissues (lower leg muscles) using a commercially available polarographic oxygen sensor system (PO2-Histograph model 6650; Eppendorf). These animals were not subjected to imaging studies. Measurements were obtained in anesthetized animals bearing tumors of the same size range as used for imaging studies (EMT6 tumors, 0.9 ± 0.2 g; AR42J tumors, 0.8 ± 0.2 g). After inserting the tip of the oxygen sensor below the surface of the tumor and allowed it to equilibrate, 6 tracks were measured (tracks averaged 6 mm in length), resulting in 120 single PO2 measurements per tumor. Measurements were taken after moving the needle forward automatically in 0.5-mm steps (a rapid forward movement of 0.7 mm was followed by a backward movement of 0.2 mm to minimize compression effects caused by forward motion of the needle). The results are expressed as the geometric mean and the 95% confidence limits of each tissue histogram. Individual PO2 results were then pooled into frequency histograms summarizing the measurements for each tumor model as percentages of relative frequency of the tissue PO2 in mm Hg.
Biodistribution Studies
Biodistribution studies were performed in EMT6 tumor-bearing BALB/c mice and AR42J and AR431 Swiss nude mice. After intravenous injection of either 18F-FMISO or 18F-FAZA, mice were awake until sacrifice. Animals were euthanized in CO2 gas and rapidly dissected 10, 60, and 180 min after dosing with either 18F-FMISO or 18F-FAZA by intravenous injection. Biodistribution measurements were evaluated in all tumors and in various organs, including blood, heart, lung, liver, spleen, kidney, brain, muscle, bone, skin, stomach, small intestine, and large intestine. The selected tissues and organs were harvested and weighed. HPLC (RP-18, gradient H2O/ethanol) and thin-layer chromatography (TLC) (silica gel, ethyl acetate) were performed to analyze radioactivity in urine. The tumor-to-blood ratio (T/Bl ratio) and tumor-to-muscle ratio (T/M ratio) were considered for statistical analysis. The radioactivity in the tissue was measured using a 
-counter (1480 Wizard automatic -counter; Wallac OY). Results are expressed as the percentage of injected dose per gram of tissue (%ID/g).
Animal PET
To determine the hypoxia specificity of the 18F-FAZA uptake, serial small animal PET was performed using the Munich prototype animal PET scanner (MADPET) (31). The experiments were performed in 13 A431 tumor-bearing Swiss nude mice, randomly selecting whether they breathed pure oxygen (100% O2) or room air (21% O2) during the first day and selecting the other atmosphere on the following day. Animals breathing oxygen were kept in a custom-built acrylic container, flushed with pure oxygen (5 L/min), beginning 1 h before tracer injection and continuing to the end of the PET study. Except for a period of approximately 1–3 min for tracer injection (ether sedation), the oxygen atmosphere was never compromised.
Two hours after intravenous injection of 
20 MBq 18F-FAZA, animals were anesthetized and positioned prone inside the animal PET tomograph. PET studies were performed on anesthetized animals (intraperitoneal injection of ketamine, 70 mg/kg [Pharmacia], and xylazine, 7 mg/kg [Bayer]).
The animal PET scanner has a reconstructed image resolution of 2.5 mm (full width at half maximum) in a transaxial field of view of 75 mm and a slice thickness of 2 mm (32). Depending on the tumor size, 2–6 transaxial slices through the tumor region (10 min per slice) were measured.
PET Data Analysis
After correction for dead time and random coincidences, PET data were reconstructed using a statistical reconstruction method (ordered-subset expectation maximization) including a scanner-specific system response matrix based on Monte Carlo simulation of the complete system (33). Standardized regions of interest (ROIs) were placed over the maximum activity in the tumor in 18F-FAZA-PET images. The slice with the highest radioactivity concentration within the tumor was visually identified and a circular ROI with a diameter of 5 mm was automatically placed in the area with the maximum activity concentration, corresponding to the tumor. An additional circular ROI with a diameter of 15 mm was placed in the adjacent background of the animal thorax as the reference.
Assessment of tracer distribution in tumor tissue was expressed as the T/Bk ratio, dividing the mean activity within the ROI of the tumor by the mean activity within the background ROI.
Autoradiography
18F-FAZA autoradiography studies were performed in 17 EMT6 tumors and corresponding muscle tissue as the reference. The animals were either breathing room air (8 tumors) or carbogen (95% O2 and 5% CO2) (9 tumors) until sacrifice 3 h after 18F-FAZA injection. The tissues were quickly frozen (–70°C) and cut into 20-µm sections (Cryo Polycut; Leica). Radioactivity was determined 4–5 h after injection using a Phosphor Imager 445 SI (Molecular Dynamics, Amersham). Distribution of radioactivity in tumor and in muscle was calculated using eliptic ROIs. T/M ratios from mice breathing room air or carbogen were compared quantitatively.
In addition, to visualize the whole-body distribution of 18F-FAZA activity in 2 mice bearing AR42J tumors, whole-body autoradiography was performed. After sacrifice, animals were quickly frozen (–70°C) and cut into 20-µm sections (Crystat Microtome CM 3500; Leica). Radioactivity was determined 4–5 h after injection using the same Phosphor Imager as used for selective autoradiography.
Statistics
Results are expressed as mean values of parameters ± SD. Parameters were compared by means of 1-way ANOVA, including tests for homogeneity of group variances using the Bartlett test, selecting a conservative significance level of P 
0.1. In case of inhomogeneous group variances, data were compared using a t test. If variances were not equally distributed, group differences were assessed using the Wilcoxon rank sign test. Parameters were correlated using a nonparametric test (Spearman rank correlation). P < 0.05 was considered statistically significant. Statistical tests were performed with the JMP version 3.2 statistical software package (SAS).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
90% (n = 2) of activity and at 60 min after injection 73% (n = 1) of activity were due to unchanged 18F-FAZA.
|
|
Animal PET
In all 18F-FAZA-PET images obtained from A431 tumor-bearing Swiss nude mice, tracer uptake was clearly visible in the tumor. Serial 18F-FAZA-PET scans were performed in 13 A431 tumor-bearing Swiss nude mice that breathed 2 different atmospheres (21% O2 and 100% O2). Figure 2 shows 2 representative animals with their consecutive PET scans. In all PET studies, the T/Bk ratio was significantly elevated in mice breathing room air compared with that of those breathing pure oxygen (T/Bk ratio: 7.3 ± 2.3 and 4.2 ± 1.2, respectively; P < 0.001) (Fig. 3).
|
|
PET measurements were confirmed comparing the T/Bk ratio derived from PET with the T/M ratio derived from direct tissue assays. Figure 4 shows the relationship between the T/M and T/Bk ratios. Regression analysis revealed a significant linear correlation between both measurements (regression line y = 0.89 + 0.75x; adjusted r2 = 0.57; P < 0.01; dotted lines, 95% confidence intervals).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Compared with 18F-FMISO, 18F-FAZA displays a lower octanol:water partition coefficient (log P = 1.1), indicating the potential for both rapid diffusion through tissue and faster renal excretion (35). In addition, the significantly lower T/Bl ratio resulting from 18F-FAZA administration, which is related to either renal or hepatobiliary excretion, leads to a lower radiation burden and to a favorable imaging result compared with 18F-FMISO. In contrast, Sorger et al. found that 18F-FMISO displayed a slightly higher T/Bk ratio compared with 18F-FAZA in a Walker 256 rat sarcoma model (36). These conflicting results might be explained by species-related differences as well as the fact that this study was conducted with large, partially necrotic tumors, making direct comparison with the current study difficult.
We verified tumor tissue hypoxia using oxygen electrode measurements in 2 of the 3 investigated animal models (EMT6/BALB/c and AR42J/nude mice). Although this device has been established as the gold standard for tissue PO2 measurements in humans and large animals (5,20), it is of limited value in small animal models. The needle electrode is moved forward stepwise by a motor, advancing the tip of the probe through the tumor tissue, thus avoiding compression artifacts and performing measurements throughout a sector of tumor tissue. During measurement, the exact location of the probe’s tip cannot be monitored. Therefore, measurements from the rim of tumors are difficult to perform and the results may include PO2 values obtained from necrotic areas within the tumor tissue. In addition, tumors are traumatized by this invasive procedure, prohibiting repeated measurements. Also, at the time of PO2 measurement, certain anesthetics and sedation procedures may have profound effects on invasive probe measurements (37), whereas radionuclide investigations allow animals to metabolize after tracer delivery until imaging or sacrifice. These limitations may explain the generally poor correlation between hypoxia tracer uptake and tissue PO2 measurements in unperturbed animals (38). Considering these limitations for this study, PO2 measurements were not performed in animals undergoing further imaging.
In the present study, the oxygen dependency of 18F-FAZA uptake has been demonstrated not only by serial small animal PET but also by biodistribution and autoradiographic studies under different oxygen conditions and in different tumor models. In the autoradiographic study, a carbogen atmosphere, rather than pure oxygen, was used to counteract possible constricting effects of pure oxygen on the microvasculature. In serial PET studies, potential influences of tumor growth on tumor tissue oxygenation were avoided by randomly alternating the atmosphere breathed by test animals (e.g., first oxygen, then air the following day, or vice versa). One could argue that vasoconstriction due to breathing pure oxygen may have caused the observed effects on 18F-FAZA uptake in A431 tumors. Nevertheless, we can assume that tumor and normal tissues were subjected to possible vasoconstrictive effects to a similar degree. Therefore, vasoconstrictive effects of the oxygen atmosphere should not have greatly influenced our measurements, because only the ratio of tumor-to-reference tissue was used to describe 18F-FAZA retention. Also, a comparable reduction of the 18F-FAZA uptake was observed in autoradiographic studies using carbogen instead of pure oxygen.
Several important preconditions have to be met by bioreductive tracers to be useful for hypoxia imaging. Especially important are (a) the amount of tracer delivered, (b) the time course of tracer transport throughout the tumor and across the tumor cell membrane, (c) the specific tissue retention mechanism, (d) the amount of unspecific metabolites in the circulation, and (e) the rate of excretion. These parameters mainly influence the target-to-background ratio that ultimately determines the ability to detect tissue hypoxia (sensitivity) and the range of oxygen concentration that defines the radiobiologic oxygen effect—in other words, the prediction of tumor response to therapy (specificity). The fact that tissue hypoxia is generally related to a significant decrease in tissue perfusion is unfortunate for imaging purposes. Therefore, tracer delivery can be critically impaired due to decreased blood flow. It has been shown that the reduction of blood flow below 10% of normal values critically reduces tracer delivery (11). However, it has been reported that 125I-IAZA, which is more lipophilic than 18F-FAZA, undergoes perfusion into areas not delineated by the perfusion marker 99mTc-hexamethylpropyleneamine oxime in an animal stroke model, indicating good penetration into poorly perfused tissue (39). In this model, 125I-IAZA uptake became visible at a threshold blood flow of 34%. Magnetic resonance-derived apparent diffusion coefficient maps identified 4 distinct regions, of which the region with the lowest apparent diffusion coefficient correlated with the highest 125I-IAZA uptake. Clearly, hypoxia tracers can be used successfully only as long as tracer delivery to tissue is maintained; thus, it is important to have a highly diffusible tracer with little tendency to partition into lipophilic membranes or adipose tissue. The tracer should easily cross the blood capillary membrane, preferably without using membrane transporter systems, which otherwise might complicate the interpretation of tracer studies. Most importantly, an oxygen-specific retention mechanism should determine the amount of tracer that is temporarily or permanently trapped in hypoxic tissues. This implies that only viable cells that are hypoxic will retain hypoxia tracers. Ideally, labeled metabolites of hypoxia tracers should not be found in the circulation at the time of imaging. None of the present hypoxia tracers available completely meets all these preconditions.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
For correspondence contact: Morand Piert, MD, Division of Nuclear Medicine, Department of Radiology, University of Michigan Health System, University Hospital B1G505C, 1500 E. Medical Center Dr., Ann Arbor, MI 48109-0028.
E-mail: mpiert{at}umich.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Related articles in JNM:
This article has been cited by other articles:
|
|
 |
|
  P. T. Winnard Jr., A. P. Pathak, S. Dhara, S. Y. Cho, V. Raman, and M. G. Pomper Molecular Imaging of Metastatic Potential J. Nucl. Med., June 1, 2008; 49(Suppl_2): 96S - 112S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Krohn, J. M. Link, and R. P. Mason Molecular Imaging of Hypoxia J. Nucl. Med., June 1, 2008; 49(Suppl_2): 129S - 148S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Takasawa, R. R. Moustafa, and J.-C. Baron Applications of Nitroimidazole In Vivo Hypoxia Imaging in Ischemic Stroke Stroke, May 1, 2008; 39(5): 1629 - 1637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Langen and S. M. Eschmann Correlative Imaging of Hypoxia and Angiogenesis in Oncology J. Nucl. Med., April 1, 2008; 49(4): 515 - 516. [Full Text] [PDF]
|
|
|
|
|
|
M. Picchio, R. Beck, R. Haubner, S. Seidl, H.-J. Machulla, T. D. Johnson, H.-J. Wester, G. Reischl, M. Schwaiger, and M. Piert Intratumoral Spatial Distribution of Hypoxia and Angiogenesis Assessed by 18F-FAZA and 125I-Gluco-RGD Autoradiography J. Nucl. Med., April 1, 2008; 49(4): 597 - 605. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Wester Nuclear Imaging Probes: from Bench to Bedside Clin. Cancer Res., June 15, 2007; 13(12): 3470 - 3481. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Beck, B. Roper, J. M. Carlsen, M. C. Huisman, J. A. Lebschi, N. Andratschke, M. Picchio, M. Souvatzoglou, H.-J. Machulla, and M. Piert Pretreatment 18F-FAZA PET Predicts Success of Hypoxia-Directed Radiochemotherapy Using Tirapazamine J. Nucl. Med., June 1, 2007; 48(6): 973 - 980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Veit-Haibach, C. A. Kuehle, T. Beyer, H. Stergar, H. Kuehl, J. Schmidt, G. Borsch, G. Dahmen, J. Barkhausen, A. Bockisch, et al. Diagnostic Accuracy of Colorectal Cancer Staging With Whole-Body PET/CT Colonography JAMA, December 6, 2006; 296(21): 2590 - 2600. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A-L Grosu, M Piert, and M Molls Experience of PET for target localisation in radiation oncology Br. J. Radiol., November 1, 2005; Supplement_28(1): 18 - 32. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Solomon, D. Binns, P. Roselt, L. I. Weibe, G. A. McArthur, C. Cullinane, and R. J. Hicks Modulation of intratumoral hypoxia by the epidermal growth factor receptor inhibitor gefitinib detected using small animal PET imaging Mol. Cancer Ther., September 1, 2005; 4(9): 1417 - 1422. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| JOURNAL OF NUCLEAR MEDICINE TECHNOLOGY | THE JOURNAL OF NUCLEAR MEDICINE |
