Influence of animal heating on PET imaging quantification and kinetics: results on the biodistribution of 18 F-tetrafluoroborate and 18 F-FDG in mice

Size: px

Start display at page:

Download "Influence of animal heating on PET imaging quantification and kinetics: results on the biodistribution of 18 F-tetrafluoroborate and 18 F-FDG in mice"

Ariel Andrews
6 years ago
Views:

1 Journal of Nuclear Medicine, published on December 15, 2016 as doi: /jnumed Influence of animal heating on PET imaging quantification and kinetics: results on the biodistribution of 18 F-tetrafluoroborate and 18 F-FDG in mice Christian GOETZ 1,2,3, Matthias PODEIN 3, Friederike BRAUN 3, Wolfgang A. WEBER 4, Philippe CHOQUET 1,2, André CONSTANTINESCO 1,2,Michael MIX 3 1 UF6237 Imagerie Préclinique, Pôle d imagerie, CHU Hautepierre, Hôpitaux Universitaires de Strasbourg, Strasbourg, France 2 ICube, Département Mécanique, CNRS, Strasbourg, France 3 Department of Nuclear Medicine, Medical Center University of Freiburg, Faculty of Medicine, University of Freiburg, Germany 4 Molecular Imaging and Therapy Service, Memorial Sloan-Kettering Cancer Center, New York, USA Disclaimer: none Corresponding author: Prof. André CONSTANTINESCO Hôpitaux Universitaires de Strasbourg Service de Biophysique et Médecine Nucléaire 1 Avenue Molière Strasbourg Telephone: Fax: aconstant@unistra.fr First author: Dr. Christian GOETZ UNIVERSITÄTSKLINIKUM FREIBURG Klinik für Nuklearmedizin Hugstetterstraße Freiburg Telephone: Fax: christian.goetz@uniklinik-freiburg.de Word count: 2472

2 ABSTRACT Varying environmental conditions under anesthesia may lead to unstable homeostatic conditions in rodents and thus alter kinetics. This study evaluated the impact of varying heating conditions on PET imaging quantification. Methods: Two groups of six adult female balb-c nude mice with subcutaneously implanted tumors underwent micropet imaging after injection of 18 F-tetrafluoroborate and 18 F-fluordeoxyglucose respectively. Dynamic scans were acquired under optimal and sub-optimal heating conditions. Time-activity curves were analyzed to calculate uptake and washout time constants. Results: Optimal animal heating showed stable heart rates during acquisition (515 ± 35 bpm) whereas sub-optimal heating led to lower heart rates with higher standard deviation (470 ± 84 bpm). Both uptake and washout time constants were faster (p < 0.01) in animals maintained under optimal heating conditions. Conclusion: Although the differences in heart rate were slight, optimal heating yields significantly faster uptake and washout kinetics than sub-optimal heating in all organs for both tracers. Key words: Homeostasis, PET kinetics, tracer biodistribution, xenografts

3 INTRODUCTION Pre-clinical functional imaging studies in mice have been widely used for translational research (1-4). Nevertheless, the physiological parameters of mice during the imaging studies are still poorly standardized. These parameters can be affected by fasting (5-7) and by anesthesia during or following the injection of the tracer (8) or during the image acquisition procedures (6, 7). Differences in animal handling may therefore lead to considerable differences among the results reported by different investigators (7, 9, 10). A controlled and standardized environment (10, 11), including consideration of circadian rhythms (5, 9), are known to lead to more reproducible functional imaging results in mice (10, 11). On the other hand, animal exposure to varying conditions - especially under anesthesia - is deleterious to their homeostasis (6, 7) and will induce poor reproducibility of experimental outcomes (7, 8). Homeostasis can be maintained, however, if the underlying regulatory functions under experimental conditions are controlled in such a way that such functions approximate those of the normal, awake animal. As heart rate, respiration frequency and body temperature can be monitored in mice (12), we aimed to study the influence of the ambient/heating temperature on these physiological rates. As changes in both heart rates and mouse body temperature might also influence the distribution of labelled tracers as well as their kinetics, we aimed then to evaluate throughout this study the impact of modifying heating conditions on PET imaging quantification for two tracers, 18 F labelled tetrafluoroborate ( 18 F -TFB) and 18 F labelled fluordeoxyglucose ( 18 F -FDG). MATERIALS AND METHODS Animals care, preparation and use were carried out in accordance with the European legislation on laboratory animals and animal studies. Impact of ambient temperature on physiologic parameters In order to study the impact of ambient temperature on heart rates, respiratory rate and body temperature we first studied three groups of 6 normal female BALB/c nude mice (body mass 22.3 ± 0.7 g). Animals were maintained under continuous anesthesia inside a dedicated animal holder (Minerve imaging cell, Minerve SAS, Esternay,

4 France) and their physiological parameters were continuously monitored during the whole procedure. Heart rate was measured using three carbon tube electrodes positioned on the animal extremities (12) and body temperature assessed by a dedicated intra-rectal probe. Anesthesia gas mix was kept constant to avoid any influence of isoflurane concentration on animals homeostatic capabilities. All groups were maintained under a 1.5% isoflurane and oxygen mix. All animals were warmed at the same temperature for a 15-minute delay before any recording to ensure stabilization of physiological rates. After this initial delay period, the ambient temperature in the holder was set either to 33, 35 or 37 C for the first, second or third group, respectively, and individual heart rates and body temperature were measured for a 60-minute observation period. Results from the follow up are presented and discussed in the following sections. From these first monitoring observations and as a preamble to the kinetic measurements we defined and retained for the forthcoming imaging procedure, two sets of heating conditions. Optimal heating conditions were defined by a holder maintained at a constant temperature of 35 C, sub-optimal conditions by a temperature of 33 C. These conditions were applied successively to our implanted animal groups during the imaging procedures. Higher temperatures for the holder (37 C) lead to overheating attested by an increase in the measured mouse body temperatures (results section) as wells as unstable anesthesia with uncontrolled limb movements and undesired awaking. These heating conditions are also discussed in the following sections but were not therefore considered further for the imaging procedures. Imaging of animals under different physiological conditions A pool of 12 adult female BALB/c nude mice (body mass 18.7 ± 0.3g) was included and selected for tumor implantation. Wild-type anaplastic thyroid cancer cells and human sodium iodine symporter (hnis) transfected cancer cells xenografts were implanted subcutaneously in all animals. Implanted cell preparations were obtained and prepared according to Gholami and al. (13). Transfected xenografts were implanted between the left shoulder and flank of the animal and wild-type xenografts between the right shoulder and flank of the same animal (14). Imaging protocol

5 After 4 days of tumor growth (both implanted tumors sizes were 5-7 mm), animals were prepared for imaging procedures. Animals were provided with food and water ad libitum prior to imaging with both tracers. Mice were maintained under anesthesia inside the same animal holder attached to the micro-pet device (R4 micropet, Concorde Microsystems, US). The mice were kept warm by the holder circulating air and stabilized before tracer administration. Tracers, 18 F-TFB and 18 F-FDG, were injected separately directly in a tail vein inside the holder just before the beginning of imaging acquisition. 18 F-TFB was produced in our group according to Jauregui et al. (15). Each animal underwent four tracer injection followed by dynamic PET acquisitions, two with 18 F-FDG on day one and two for 18 F-TFB on day 2. A four-hour free interval was set between the two successive acquisitions at the same day. During this interval and during the night, mice had food and water ad libitum. Circadian rhythm is known to influence homeostatic and metabolic regulation systems and might therefore influence our tracer uptake (5, 9) depending on the time of injection and on the time of the last food intake of the animal. Tumor growth between the first and the last acquisition ( hours) might also influence the tracer uptake and confound intra-individual comparisons. To limit the influence of these circadian effects (including feeding habits) as well as the tumoral growth, four groups of three animals were constituted and for each group, the imaging sequence was cycled between optimal and sub-optimal heating conditions for both tracers as well as the time of injection to avoid any systematic influence or risk of systematic error. Mice were administered twice with 3.1 ± 0.2 MBq 18 F-TFB and twice with 3.2 ± 0.4 MBq 18 F-FDG. Dynamic PET images were acquired over the first 45 minutes after injection. Optimal or sub-optimal heating conditions for the holder were kept stable during the whole acquisition process. Mouse heart rates were monitored and recorded throughout the 45 minutes. In total, 48 datasets were collected for the 12 mice with bilaterally implanted tumors. Image reconstruction and data collection Ninety successive frames were collected for each dataset during the first 45 minutes (30 seconds per frame) with a single bed position covering the whole mouse body (full axial field of view was 7.2 cm). Data reconstruction used the vendor s OSEM2d algorithm (R4 micropet, Concorde Microsystems, US) with 16 subsets and 4 iterations. 18 F decay correction was applied but no attenuation correction method was selected. Final image reconstruction matrix was 128x128x95 pixels for an 11x11x8 cm field of view. After data reconstruction (3D datasets + Time) and to assess time-activity data insuring reproducible ROI positioning, a time-series of 2D-projection images was generated: ninety successive single 2D ventral to dorsal

6 projection images were calculated from each 3D frame. This ventral to dorsal projection image was preferred for region of interest positioning to ensure positioning reproducibility as well as limited overlapping of organs of interest compared to the other axis projections. Several regions of interest (ROI) were manually drawn on the ventral to dorsal projection frames and used to derive time-activity data. ROIs were placed over the thyroid, stomach, implanted tumors and blood compartment (heart ROI) for 18 F-TFB images analysis (Fig. 1 - top). ROIs were placed over the liver, brain, heart and implanted tumors for 18 F-FDG images analysis (Fig. 1 - bottom). The sizes and shapes of the ROIs were the same for both heating conditions. Curve fitting and kinetic measurements Raw time-activity curves were obtained from the ROIs for each animal for both optimal and sub-optimal heating acquisitions and for both injected tracers. Image-derived time-activity data for organs and tumor were analyzed using bi-exponential models: tracer concentration =A ( - ) + B where A and B are fitting constants with units of tracer concentration and λ washout and λ uptake, respectively, the washout and uptake time constants with units of minutes -1. Time-activity curves raw points collected from blood pool compartments are fitted by a bi-exponential decreasing function: tracer concentration =A ( + ) + B, where A and B are fitting constants and λ fast washout and λ washout, respectively, the initial fast apparent washout and the slow late washout time constants. Dedicated modeling software was used for the determination of these uptake and washout time constants (profit, Quantumsoft, Switzerland). Statistical analysis A paired Student s t-test was used for direct comparison of uptake and washout time constants for each compartment of the model comparing optimal and sub-optimal heating regimes. Values of p < 0.01 were considered significant. RESULTS

7 Optimal and sub-optimal heating conditions were determined to correspond to circulating heating air maintained at a temperature of 35 C and 33 C, respectively. Heart rates and body temperature were continuously monitored over the 60-minutes observation period for all mice. Fig. 2 illustrates the mean evolution of both measured parameters for all three groups. Mean heart rates after the first 15 minutes were 482 ± 45 bpm and mean body temperature 36.9 ± 0.1 C for the first group, 487 ± 40 bpm and mean body temperature 37.0 ± 0.2 C for the second group and 483 ± 53 bpm and mean body temperature 36.9 ± 0.1 C for the third and last group. No statistically significant difference among groups was found. The first of our groups showed a continuous and slow decrease in both heart rates and body temperature over the 60 minutes of observation. Final heart rate was 390 ± 120 bpm and final body temperature 35.9 ± 0.6 C. The second group showed much better stability during recording: final heart rate was 489 ± 35 bpm and final body temperature 36.9 ± 0.3 C. The third and last group was associated with a continuous increase in both heart rate and temperature: final heart rate was 584 ± 62 bpm and final body temperature 37.4 ± 0.7 C. According to these preliminary measurements and for our specific animal holding system, homeostasis under anesthesia was best conserved for a circulating air temperature set to 35 C (mentioned as optimal heating). Suboptimal heating of the animal (33 C) led to a progressive decrease in body temperature measured with a rectal probe and was accompanied by a reduction in heart rate. After 60 minutes, the mean difference in body temperature and heart rate between the groups was 1 C and 19%, respectively. Excessive heating (holder temperature maintained to 37 C) led to an overall body heating of 0.5 C and a heart rate increase of more than 20%. For this last group, anesthesia was also harder to control; measurements had to be repeated for three mice that awoke in the last 15 minutes of the procedure. These last heating conditions were therefore not considered further for the imaging procedures. Imaging and kinetics measurements Time constants and mean heart rates measurements (mean values over all 12 animals) are summarized in Tables 1 and 2, respectively, for 18 F- TFB and 18 F-FDG tracers. After bolus intravenous injection, 18 F-TFB tracer is rapidly distributed over all body compartments. Fig. 3 illustrates 18 F-TFB time-activity curves for a single mouse. Fitting and time constant calculation led to an overall faster uptake and washout of the tracer for optimal heating conditions.

8 Tracer wash-out from blood was about 5 times faster for optimal heating (λ washout = ± min-1) compared to the sub-optimal heating condition (λ washout = ± min-1). This difference was significant (p < 0.01, n=12). Paired comparison showed for the whole mouse population significant faster uptakes (p < 0.01, n=12) for thyroid, stomach and both tumors as well as faster washout (p < 0.01, n=12) for these compartments. The measured data and calculational results for 18 F-FDG tracer were similar to those for 18 F-TFB (Fig. 4 and Table 2). Uptake was faster for the brain, heart muscle and both implanted tumors (p < 0.01, n=12) under optimal heating conditions. Washout from the same organs was also significantly faster under optimal heating (p < 0.01, n=12). Maximum 18 F-TFB and 18 F-FDG tracer uptake was not significantly different between optimal and sub-optimal conditions as illustrated in figures 3 and 4, respectively, but the respective fitted curves showed a systematic additional delay to maximum concentration (shifted time to maximum) for the animals encountering sub-optimal heating - this delay is related to the associated overall slower measured kinetics. Uptake time constants from the liver compartment were not assessable to calculation - the selected 30-second frame duration rate we used was too coarse to reliably assess the time course of initial radiotracer uptake in the liver. All mice survived the imaging procedures; continuous heart rate monitoring during the imaging process allowed calculation of mean rates for both heating conditions. A slight but significant difference (p < 0.01) between optimal and sub-optimal heating conditions was observed (Tables 1 and 2). Optimal heating conditions resulted in higher heart rates during the imaging procedures as well as reduced standard deviations (515 ± 35 bpm for 18 F-TFB and 537 ± 55 bpm for 18 F-FDG) compared to sub-optimal heating conditions (470 ± 84 bpm for 18 F-TFB and 442 ± 96 bpm for 18 F-FDG). DISCUSSION AND CONCLUSION Different heating conditions during anesthesia led to slight but statistically significant differences in heart rate during PET imaging as well as a larger standard deviation associated with sub-optimal heating. For both 18 F-TFB and 18 F-FDG, optimal heating yielding faster uptake and clearance kinetics in all organs evaluated.

9 This highlights the influence of even small differences in animal heating temperature on tracer distribution. In agreement with previously published work (6, 7, 9, 10), a strict control of all homeostasis influencing parameters (body temperature, fasting, anesthesia conditions) is mandatory to insure reproducibility of experimental results.

10 REFERENCES 1. Schelbert HR, Hoh CK, Royal HD, et al. Procedure guideline for tumor imaging using fluorine-18-fdg. Society of Nuclear Medicine. J Nucl Med. 1998;39: Gambhir SS. Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer. 2002;2: Phelps ME. Positron emission tomography provides molecular imaging of biological processes. Proc Natl Acad Sci USA. 2000;97: Cherry SR, Gambhir SS. (2001). Use of positron emission tomography in animal research. ILAR J. 2001;42: Hedrich HJ, Bullock G. The Laboratory Mouse, The Handbook of Experimental Animals. Amsterdam: Elsevier Science & Technology; Lee KH, Ko BH, Paik JY, et al. Effects of anesthetic agents and fasting duration on 18F-FDG biodistribution and insulin levels in tumor-bearing mice. J Nucl Med. 2005;46: Fueger BJ, Czernin J, Hildebrandt I, et al. Impact of animal handling on the results of 18F-FDG PET studies in mice. J Nucl Med. 2006;47: Toyama H, Ichise M, Liow JS, et al. Evaluation of anesthesia effects on [18F]FDG uptake in mouse brain and heart using small animal PET. Nucl Med Biol. 2004;31: Hildebrandt IJ, Su H, Weber WA. Anesthesia and other considerations for in vivo imaging of small animals. ILAR J. 2008;49:17-26.

11 10. Richter SH, Garner JP, Würbel H. Environmental standardization: cure or cause of poor reproducibil- ity in animal experiments? Nat Methods. 2009;6: Tseng JR, Kang KW, Dandekar M, et al. Preclinical efficacy of the c-met inhibitor CE in a U87 MG mouse xenograft model evaluated by 18F-FDG small-animal PET. J Nucl Med. 2008;49: Choquet P, Goetz C, Aubertin G, et al. Carbon tube electrodes for electrocardiography-gated cardiac multimodality imaging in mice. J Am Assoc Lab Anim Sci. 2011;50: Gholami S, Haddad D, Chen CH, et al. Novel therapy for anaplastic thyroid carcinoma cells using an oncolytic vaccinia virus carrying the human sodium iodide symporter. Surgery. 2011;150: Harzmann S, Braun F, Zakhnini A, et al. Implementation of Cascade Gamma and Positron Range Cor- rections for I-124 Small Animal PET. IEEE Trans. Nucl. Sci. 2014;61: Jauregui-Osoro M, Sunassee K, Weeks AJ, et al. Synthesis and biological evaluation of [(18)F]tetrafluoroborate: a PET imaging agent for thyroid disease and reporter gene imaging of the sodi- um/iodide symporter. Eur J Nucl Med Mol Imaging. 2010;37:

Figures - legends FIGURE 1. 18 F-TFB (upper part) and 18 F-FDG (lower part) dynamic image acquisitions. After each tracer injection, full sets of 90 images were acquired over 45 minutes.

The right upper and lower parts of the figure show a single 45 minutes projection. The circular regions of interest used to time-activity curves extraction are plotted. 18 F-TFB (upper part): 1.

12 Figures - legends FIGURE F-TFB (upper part) and 18 F-FDG (lower part) dynamic image acquisitions. After each tracer injection, full sets of 90 images were acquired over 45 minutes. The left part of each sub-figure displays optimal and sub-optimal heating acquisitions of the same animal. 9 reformatted frames are presented (5 minutes per frame) as ventral to dorsal projections. The right upper and lower parts of the figure show a single 45 minutes projection. The circular regions of interest used to time-activity curves extraction are plotted. 18 F-TFB (upper part): 1. Thyroid gland 2. Heart (blood pool) 3. Right implanted tumor 4. Left tumor 5. Stomach area and 6. Bladder; 18 F-FDG (lower): 1. Brain area 2. Right implanted tumor 3. Liver area 4. Heart 5. Left tumor 6. Kidney and 7. Bladder.

FIGURE 2. Stability of mouse cardiac frequencies and body temperature. Animals were positioned under anesthesia inside the heated animal holder.

13 FIGURE 2. Stability of mouse cardiac frequencies and body temperature. Animals were positioned under anesthesia inside the heated animal holder. After an initial 15 minutes delay to allow physiological parameters to stabilize under anesthesia, initial heating conditions were shifted to 33 C (sub-optimal, left column), 35 C (optimal, central column) and 37 C (right column) temperatures. Cardiac frequencies and mouse body temperature were recorded continuously for 60 minutes. Mean values obtained from 6 different animals are represented (solid lines) as well as the standard deviation (light grey areas).

14 FIGURE 3. Time-activity curves and fitted kinetics for the 18F-TFB tracer. Measurements from the different regions of interest are displayed for both optimal (white squares) and sub-optimal acquisitions (solid circles). A. Evolution of blood pool concentration, B. Thyroid tracer concentration, C. Left tumor tracer concentration, D. Right tumor tracer concentration, E. Stomach area tracer concentration.

15 FIGURE 4. Time-activity curves and fitted kinetics for the 18F-FDG tracer. Measurements from the different regions of interest are displayed for both optimal (white squares) and sub-optimal acquisitions (solid circles). A. Evolution of brain concentration, B. Heart area tracer concentration, C. Left tumor tracer concentration, D. Right tumor tracer concentration, E. Liver concentration.

16 Tables Table 1 Mean uptake and washout time constants and mean heart rates measured after 18 F-TFB tracer injection. Comparison of optimal and sub-optimal heating conditions is applied for each animal and p values are given after application of paired Student s t-test. Optimal heating Sub-optimal heating Blood pool washout ± min ± min -1 p < Thyroïd area uptake washout ± min ± min ± min ± min -1 p = p = Left tumor uptake washout ± min ± min ± min ± min -1 p = p = Right tumor uptake washout ± min ± min ± min ± min -1 p = p = Stomach uptake washout ± min ± min ± min ± min -1 p = p = mean heart rate during acq. 515 ± 35 bpm 470 ± 84 bpm p =

17 Table 2 Mean uptake and washout time constants and mean heart rates measured after 18 F-FDG tracer injection. Optimal heating Sub-optimal heating Brain area uptake washout ± min ± min ± min ± min -1 p = p = Heart muscle uptake washout ± min ± min -1 0,100 ± min -1 0,000 ± min -1 p = p = Liver washout ± min ± min -1 p = Left tumor uptake washout ± min ± min ± min ± min -1 p = p = Right tumor uptake washout ± min ± min ± min ± min -1 p = p = mean heart rate during acq. 537 ± 55 bpm 442 ± 96 bpm p =

Preclinical Imaging Center

Preclinical Imaging Center micropet-ct micropet microct Autoradiograph Photograph Optical Physiology matters when imaging metabolism: temperature, fasting, stress, uptake time. Positioning and reproducibility