Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The influence of tumor oxygenation on 18F-FDG (Fluorine-18 Deoxyglucose) uptake: A mouse study using positron emission tomography (PET)

The influence of tumor oxygenation on 18F-FDG (Fluorine-18 Deoxyglucose) uptake: A mouse study... Background: This study investigated whether changing a tumor's oxygenation would alter tumor metabolism, and thus uptake of F-FDG (fluorine-18 deoxyglucose), a marker for glucose metabolism using positron emission tomography (PET). Results: Tumor-bearing mice (squamous cell carcinoma) maintained at 37°C were studied while breathing either normal air or carbogen (95% O , 5% CO ), known to significantly oxygenate 2 2 tumors. Tumor activity was measured within an automatically determined volume of interest (VOI). Activity was corrected for the arterial input function as estimated from image and blood- derived data. Tumor FDG uptake was initially evaluated for tumor-bearing animals breathing only air (2 animals) or only carbogen (2 animals). Subsequently, 5 animals were studied using two sequential F-FDG injections administered to the same tumor-bearing mouse, 60 min apart; the first injection on one gas (air or carbogen) and the second on the other gas. When examining the entire tumor VOI, there was no significant difference of F-FDG uptake between mice breathing either air or carbogen (i.e. air/carbogen ratio near unity). However, when only the highest F-FDG uptake regions of the tumor were considered (small VOIs), there was a modest (21%), but significant increase in the air/carbogen ratio suggesting that in these potentially most hypoxic regions of the tumor, F-FDG uptake and hence glucose metabolism, may be reduced by increasing tumor oxygenation. Conclusion: Tumor F-FDG uptake may be reduced by increases in tumor oxygenation and thus may provide a means to further enhance F-FDG functional imaging. Page 1 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 through the tumor of a mouse (maintained at 37°C) Background F-FDG (fluorine-18 deoxyglucose) has become widely breathing air and a mouse breathing carbogen. Figure 1, used as a radiolabeled marker for positron emission tom- bottom, shows the time activity curve obtained from the ography (PET) imaging of solid tumors. In some regions VOIs drawn around the entire tumor for a tumor-bearing of the body, the predictive power for identifying cancer animal breathing air. The ordinate of the time activity 18 18 using F-FDG approaches 95% [1,2]. curve is the relative F-FDG activity concentration in the tumor per mCi injected dose. Note that the decay cor- While altered glucose metabolism is a unique feature of rected FDG uptake curve is quite flat from 60 min until the neoplastic growth, there are other factors associated with end of the study (130 min). This was true for all four mice the tumor micro-environment that are in marked contrast in the pilot study. Similar time activity curves were to normal tissues. The vascular architecture in tumor tis- obtained for animals breathing carbogen (data not sue is abnormal and differs greatly from normal tissues, shown). Note also that the uptake in these two mice (one resulting in altered blood flow and the development of on air and one on carbogen) was visually similar with tumor hypoxia. The presence of tumor hypoxia is thought considerable heterogeneity of F-FDG uptake across the to represent a barrier for effective cancer treatment for tumor. Air versus carbogen comparisons could not be both radiation and chemotherapy [3-6]. Identifying made for these animals however, because data for the patients whose tumors contain hypoxic areas may there- mice in the pilot studies could not be corrected for possi- fore have an important role in tumor prognosis as well as ble differences in the arterial input function. In addition, treatment approach and outcome. Currently, the ability to F-FDG uptake would presumably have mouse-to-mouse identify and quantify tumor hypoxia is limited. The cur- physiologic variation since different mice were used for air rent gold standard for measuring tissue oxygen concentra- and for carbogen, potentially masking true differences in tion utilizes oxygen-sensitive electrodes, which measure uptake caused by the nature of the ventilating gas. For oxygen partial pressure (pO ) directly in tumor tissue. Given the invasive nature of this technique, it is difficult to access deep-seated tumors, and once assessed, it is dif- ficult to distinguish between measurements made in necrotic and viable regions [7]. Non-invasive techniques, such as Overhauser-enhanced magnetic resonance imag- ing (OMRI) [8] and electron paramagnetic resonance imaging (EPRI) [9,10], are being evaluated to avoid the obstacles encountered with the polarographic electrode. Since non-invasive F-FDG/PET imaging is already widely used in clinical facilities to identify malignant tis- sue, we questioned whether this imaging modality might be used to assess tumor hypoxia. Anaerobic metabolism requires much more glucose to generate the same amount of ATP as under aerobic metabolism. Given the hypoxic nature of certain tumors, regions of tumor tissue are known to resort to anaerobic glucose metabolism [11]. We hypothesize that experimentally increasing a tumor's oxygen supply might enable the tumor to metabolize more glucose aerobically, thus reducing F-FDG uptake. To test this hypothesis, we have compared tumor F-FDG uptake in tumor-bearing animals sequentially breathing air (20.9% O ) and carbogen (95% oxygen, 5% CO ). 2 2 ial slices through the breathing a Figure 1 F-FDG images ( ir and a differ top) showin tumor of ent mouse breathing ca g sa a mouse (maintained gittal, coronal, an rbogen d transax- at 37°C) Carbogen breathing has been shown to markedly increase F-FDG images (top) showing sagittal, coronal, and transax- the oxygenation status of tumors [8,12]. In addition, we ial slices through the tumor of a mouse (maintained at 37°C) present data stressing the importance of maintaining nor- breathing air and a different mouse breathing carbogen. Both mal body temperature in small animals undergoing F- color scales set to the same maximum nCi/cc/injected-dose. FDG/PET imaging. Bottom: time activity curve obtained from the VOIs drawn around the entire tumor for a tumor-bearing animal breath- Results ing air (animals breathing carbogen also resulted in similar curves, flat at late times). Figure 1 shows images from two of the pilot studies. Fig- ure 1 (top) shows sagittal, coronal, and transaxial slices Page 2 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 these reasons the second more elaborate set of studies described above were undertaken. This second set of stud- F-FDG injections in the ies involved two sequential same mouse, 60 minutes apart, the first injection on one gas (air or carbogen) and the second on the other gas. The fact that the F-FDG time activity curves from the pilot studies were flat, regardless of whether air or carbogen was used, meant that in the two injection studies, we could subtract the F-FDG activity present at the end of first 60 min from the activity following the second injection. These two injection studies permitted sequential measure- ment of air and carbogen uptake in the same tumor, and Blood- and normal physiologic cor Figure 2 image-derived input functions fo e body temperature (37° r mice sca C) nned at permitted correction for possible differences in the arterial Blood- and image-derived input functions for mice scanned at input function. normal physiologic core body temperature (37°C). Blood- derived input functions were determined using blood sam- Air to carbogen uptake ratios at normal temperature 18 ples drawn every 10 min, starting 30 min after FDG injec- (37°C) tion. Thus, the blood-derived input function curve begins at time = 30 min. Note that the blood samples always yielded For each mouse we computed the ratio of the uptake lower activity concentration than the image data due to while breathing air to the uptake while breathing carbo- background activity, as per equation 1. gen, each corrected for the area under the input function as per equation 2. A typical input function is shown in Fig- ure 2, along with the late blood sample derived input function. The measured input function was corrected even if the maximum tumor value (averaged with its clos- using the blood samples, as per equations 1–3. The ratio est neighboring pixels in all 3 dimensions) was used for of corrected tumor uptake while breathing air to corrected each tumor. We wished to visualize which pixels were tumor uptake while breathing carbogen, averaged over all causing the difference in F-FDG tumor uptake for air ver- sus carbogen breathing. Figure 4 shows F-FDG tumor 5 mice, did not differ significantly from 1.0 (ratio air/car- bogen = 1.15, p = NS, Figure 3, squares). The ratio uptake images for three different animals who received remained not significantly different from unity whether two F-FDG injections, one while breathing air and the the VOI was drawn using a 30, 40 or 50% threshold, and other while breathing carbogen. The first column shows even when manually drawn VOIs encompassing the entire tumor were used. Note from Figure 3 that the tumors of four out of five mice exhibited air/carbogen ratios, all >1, while one mouse deviated from the other 4, with a ratio lower than unity. There was no a priori reason to exclude the outlying animal, but doing so resulted in an air/carbo- gen ratio, which did differ significantly from unity (air/ carbogen = 1.22, p < 0.01). As shown in Figure 1 the SCC tumor exhibited considera- ble heterogeneity in its uptake of F-FDG. We hypothe- sized that regions of the tumor demonstrating accentuated uptake of FDG might be exceptionally hypoxic and so might respond differently to an increase in the concentration of inspired oxygen as compared to other tumor tissue. To test this hypothesis we analyzed small VOI's (90% threshold) around regions of the Air/ca VOI (squar progra the tumor (9 Figure 3 rbogen ratios of m using a es), as determined with the 3D region 0% threshold) 30% threshold, and for tumor FDG uptake fo the hottest r whol growing part of e tumor tumors demonstrating the greatest activity. Unlike the Air/carbogen ratios of tumor FDG uptake for whole tumor data from the whole tumor, the uptake (corrected for VOI (squares), as determined with the 3D region growing input function) for these small hot VOIs resulted in an air/ program using a 30% threshold, and for the hottest part of carbogen ratio that was consistent across all 5 mice, and the tumor (90% threshold). Tumor uptake ratios were cor- was significantly greater than unity (air/carbogen = 1.21, rected using derived input function data as per equation 2 nd p < 0.005, Figure 3, right). Again this result was insensitive and the 2 injection activity was corrected for residual activ- ity remaining as a result of the first injection. to the way the VOI was defined. The results held when the VOI was based on either an 80% or 90% threshold, or Page 3 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 ing point in Figure 3. Negative values are not displayed in column 3. Instead the fourth column shows the (many fewer) pixels in which carbogen exceeded air (i.e. the neg- ative of the equation used for column 3). Air to carbogen uptake ratios at low temperature (30°C) When the core temperature of the mice was maintained at 30°C, an air/carbogen tumor ratio (averaged over all mice) very much less than one was observed (mean air/ carbogen ratio = 0.47 ± 0.30; n = 2). Again, this result was insensitive to the manner of VOI definition. Figure 5 shows F-FDG uptake during carbogen and air for a low temperature mouse. It is quite clear that the tumor uptake while breathing carbogen is markedly higher than the uptake while breathing air. However, whereas the blood- 18 18 S receiving two injections of b carbogen breath 3 Figure 4 agittal r 7eathing °C throughout the study) FDG tumo air and the ing, or vice-v r uptak second afte e FDG, ers images r the a ( one whi ani fo animal was switched to mals r three mice (A, lmaintaine e the animal w d at B, C) as derived input functions from mice at normal physiologic Sagittal FDG tumor uptake images for three mice (A, B, C) temperature (37°C) were highly reproducible across mul- receiving two injections of FDG, one while the animal was tiple mice, there was a large variability between blood- breathing air and the second after the animal was switched to derived input functions in the low-temperature mice. carbogen breathing, or vice-versa (animals maintained at Thus, one must interpret the data from mice not main- 37°C throughout the study). The first column shows the rel- ative tumor F-FDG uptake for air breathing. The second tained at physiologic temperature with caution. column represents tumor FDG uptake for the animal after switching to carbogen breathing (residual F-FDG from ini- Discussion tial injection subtracted). The first and second columns are Hypoxia in human tumors can significantly influence shown with the same relative color scale. The third column treatment outcome and the aggressiveness of the tumor shows the pixels in which air uptake exceeded carbogen [13]. This finding has led to considerable interest in non- uptake, expressed as 100*(AirImage - CarbogenImage)/ invasive means to assess the extent of hypoxia in tumors (mean counts in Air tumor VOI). Ideally the denominator prior to therapy, for example with MR (BOLD) imaging would have been the air image, but due to image noise con- [14,15]. Several studies have shown that tumor cell or siderations the normalization described was done instead. tumor F-FDG uptake is associated with hypoxia [16-18]. Negative values are not displayed in column 3. Instead the fourth column shows the (many fewer) pixels in which car- Glucose transporter (GLUT) receptor proteins and hexoki- bogen exceeded air (i.e. the negative of the equation used for nase activity are elevated in tumors and thought to be column 3. Corrections were made for injected activity and responsible for increased F-FDG tumor uptake com- for residual activity. The numeric values shown on the color pared to normal tissues [19]. Hypoxia has been shown to bar apply only to columns 3 and 4. the tumor F-FDG uptake for air breathing. The second column represents tumor F-FDG uptake for the animal after switching to carbogen breathing (residual F-FDG from initial injection subtracted). From these images it is difficult to tell which pixels were causing the slight decrease in tumor F-FDG uptake for animals breathing carbogen. To clarify this, the third column shows the per- centage increase in F-FDG uptake for air breathing ani- mals versus carbogen breathing animals expressed as 18 18 and the second while Figure 5 slices through th given F-FDG images showing sagittal, c two injections of e tumor breathing carbogen F-FDG, the first wh of a mouse ma oronal, inta and transaxial ined at 30 ile breathing air °C 100* (AirImage – CarbogenImage)/(mean counts in Air F-FDG images showing sagittal, coronal, and transaxial tumor VOI). Ideally the denominator would have been slices through the tumor of a mouse maintained at 30°C the air image, but due to image noise considerations the 18 given two injections of F-FDG, the first while breathing air mean value in the air tumor VOI (30% threshold) were and the second while breathing carbogen. Both images are used instead. The data were normalized so that when the displayed to the same scale, and are corrected for injected activity and residual activity produced by the first injection. tumor VOI was placed on the percentage increase image, it gave a value equal to the value shown in the correspond- Page 4 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 be a significant factor in the over-expression of these pro- carbogen ratios for overall tumor F-FDG uptake were teins [16,17,20]. In the present study we examined not significantly different from unity. However, when F-FDG uptake region of each tumor was whether information concerning hypoxia could be gained only the highest using F-FDG and PET imaging. This was considered pos- considered (perhaps the most hypoxic regions) the hypo- sible since the hypoxic nature of most tumors forces thesis was supported. In these presumably metabolically tumor glucose metabolism to utilize primarily glycolytic active regions, there was a small (21%) but significant pathways, rather than oxidative metabolism [11]. Very increase in air/carbogen uptake, suggesting reduced F- large quantities of glucose are required to produce a given FDG uptake under conditions of better oxygenation. The amount of ATP via the glycolytic pathway compared to lack of an air/carbogen difference for the whole tumor, the oxidative pathway, which supposedly accounts for and the fairly small air/carbogen difference even for the why many tumors have such high F-FDG uptake. We hottest part of the tumor might suggest that tumors are hypothesized that by oxygenating the tumor (via breath- "programmed" to burn glucose primarily in an anaerobic ing carbogen), tumor glucose metabolism would begin to fashion, regardless of their level of oxygenation. This have a significant oxidative component, requiring consid- could be a potential survival mechanism for tissue des- erably less glucose per ATP molecule produced. Thus, bet- tined to grow in an anaerobic environment. This potential ter oxygenated tumors (i.e. tumors influenced by explanation is of course speculative. Another factor which breathing carbogen) should have less F-FDG uptake could play a role is the potential vasodilatory properties of than when in their usual, hypoxic state (i.e. while the ani- carbon dioxide, although there is conflicting data mal is breathing air). To examine this possibility, F-FDG reported in the literature [24]. uptake in tumors was examined in tumors when the mouse was breathing air versus carbogen. Changes in tumor FDG uptake (hot spots) when breath- ing carbogen might add another dimension to F-FDG The SCC tumor used in the present study is known to be functional imaging. While small, the 21% change in F- quite hypoxic as determined by previous oxygen electrode FDG uptake in small regions may be giving information measurements in our laboratory [8]. Even though the similar and perhaps complimentary to that obtained by tumor is extremely hypoxic, there was no evidence of the MRI based oxygen imaging is Blood Oxygen Level necrosis in the tumors used in the present study (~1 cm Dependent (BOLD) MRI. For this imaging technique, diameter). Carbogen breathing has been shown in numer- temporal changes in the ratios of deoxy- to oxyhemo- ous studies to increase the oxygenation status of tumor- globin can be monitored utilizing the contrast mecha- bearing mice [8,21] and in patients [22-24]. Using the nisms provided by deoxyhemoglobin to protons. With same SCCVII tumor model, tumor size, and experimental this method, it is possible to examine changes in blood conditions with respect to the maintenance to core body flow and tumor oxygenation levels in response to carbo- temperature as used in the present study, we have previ- gen breathing versus air breathing [14,15]. The enhance- ously shown that carbogen breathing increases oxygena- ment observed in BOLD MRI images is typically only a tion of the SCCVII tumor as measured by Eppendorf few percent. In contrast, the decrease in F-FDG tumor oxygen electrodes [8,8]. In this study the mean tumor pO uptake upon carbogen breathing in small regions of the values for animals breathing air was 8.2 mm Hg, which tumor in the present study was up to 21%, suggesting that increased to a value of 19.8 for animals breathing carbo- in these small regions tumor glucose metabolism is gen [8]. altered as a result of increased perfusion and oxygenation. Should this observation be correct, which will require fur- As mentioned above, the hypoxic nature of the SCC tumor ther validation, the added functional dimension to F- makes it likely that a large proportion of the tumor cells FDG imaging afforded by carbogen breathing could be would utilize glucose by anaerobic glycolysis, and so useful to clinicians assessing treatment response, particu- exhibit enhanced F-FDG uptake. Indeed, we found glu- larly for radiation treatment. cose utilization as measured by uptake of F-FDG to be significant in the SCC tumor as evidenced by the tumor Our results did not wholly agree with some previous stud- images shown in Figure 1. The FDG uptake in the tumor ies that showed whole tumor F-FDG uptake increases in was particularly striking compared to the non-tumor bear- more hypoxic conditions [25-28]. Our whole tumor ing leg where little to no uptake was observed (data not results gave an air/carbogen uptake ratio that did not dif- shown). It was anticipated that increasing the oxygenation fer significantly from unity. Only when VOIs were drawn of the tumor by carbogen breathing would lead to a shift around the hottest part of our tumors were our findings to aerobic metabolism producing much more ATP per more consistent with these previous studies. The afore- molecule of glucose, thereby decreasing the need for glu- mentioned studies either examined a different tumor type cose, and decreasing F-FDG tumor uptake. The results of from our study (C3H mammary carcinomas) [25], or this study only partially supported this hypothesis. Air/ examined the same tumor type (SCC), but in an in vitro Page 5 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 setting [26]. These differences alone could potentially at 37°C. When only the highest uptake regions of the account for the difference in our findings. Additionally, tumor were considered, however, there was a modest this result could be due to differences in experimental (21%) but significant increase in the air/carbogen ratio. method, or due to the variability in our own results. Pre- This suggests that in these potentially most hypoxic vious studies did not attempt to correct for potential dif- regions of the tumor, F-FDG uptake may be reduced by ferences in input function. The lack of standardization increases in tumor oxygenation and thus may provide a across animal PET studies, both in the means of determin- means to further enhance F-FDG functional imaging. As ing reference input functions, as well as lack of consensus a secondary finding, we note that when the mouse's tem- regarding simpler procedures such as controlling animal perature was permitted to fall to 30°C (as occurs naturally core temperature, might also explain this difference in during isoflurane anesthesia), significant alterations in result. Future experiments with a greater number of mice, F-FDG metabolism were observed, making it clear that using simultaneous blood sampling, would clarify the strict physiological controls are necessary when such results shown in Figure 3. experiments are performed. The ability to examine and quantify regions of greatest Methods change in F-FDG uptake, which is where a radiation Animals and tumors oncologist would presumably concentrate three-dimen- Female C3H mice, produced by the National Cancer Insti- sional conformal radiation dose painting, has significant tute Animal Production Area (Frederick, MD), were used implications for cancer diagnosis and treatment. Future for this study. The mice were 7–9 weeks of age at the time studies could further test this phenomenon by shifting to of experimentation and weighed between 20–30 grams. more hypoxic conditions, to see if F-FDG uptake All experiments were carried out under a protocol improves even more. approved by the National Cancer Institute Animal Care and Use Committee, and were in compliance with the As an adjunct to our study, we examined the results Guide for the Care and Use Of Laboratory Animal Resource, obtained when mice were allowed to drop core body tem- (1996) National Research Council. Tumors were grown in perature to 30°C. Past work in our laboratory has demon- the mice by a subcutaneous (s.c.) injection of a single-cell squamous cell carcinoma (SCCVII) strated that animal's core body temperature under suspension of 2 × 10 isoflurane anesthesia drops ten degrees in an average of 15 cells in the right hind leg. Tumors grew to a size of ~1.0 min [27]. We found that mice at this lower temperature cm diameter 7–10 days after injection. Tumor volumes showed significantly greater F-FDG accumulation in car- were determined prior to PET scanning, by measuring bogen than air environments. This could be due to relative three orthogonal diameters using calipers. For all imaging vasoconstriction in low temperature conditions, possibly studies ~1 cm diameter tumors were used. making F-FDG uptake flow limited in these circum- stances. Adding carbogen to this low-temperature state PET scanner may improve blood flow, because vasculature responds to PET images were obtained with the ATLAS (Advanced the high partial pressure of oxygen by vasodilation. Technology Laboratory Animal Scanner), a dedicated Although this reasoning is speculative, our low tempera- small animal PET scanner developed at NIH with an axial ture findings do point definitively to the necessity for field-of-view (FOV) of 2 cm, a transverse FOV of 6.8 cm strict control of the mice's physiology and the need for and an aperture of 8 cm [28-30]. standardization across animal PET studies. The image data acquired with the ATLAS PET scanner were One strength of the present study was the ability to meas- reconstructed using the three-dimensional ordered subset ure F-FDG uptake with air and carbogen breathing in expectation maximization (OSEM) reconstruction algo- the same tumor. Previous studies have compared uptake rithm, including a model of the system resolution [31]. across different animals. Differences in tumor heterogene- All reconstructions were performed on the NIH Biowulf ity across mice potentially confound these results. Our computer cluster, utilizing sixteen subsets and taking 10 study controlled for this confounding variable. Further iterations. Under these conditions, a spatial resolution of studies would benefit from testing multiple tumor types, 1.6 mm full-width-at-half-maximum (FWHM) was as well as human tumors. achieved. PET scan experimental set-up Conclusion The ratio of F-FDG uptake in SCC tumors while breath- The mice were fasted at least 3 hours prior to injection of ing air to that while breathing carbogen, did not differ sig- F-FDG. Each mouse was anesthetized with isoflurane nificantly from unity when the entire tumor was (induction; 2.0%, maintenance; 1.5%) carried by air at considered, and when the animal's temperature was kept 750 ml/min and delivered via a nosepiece. An anesthetic Page 6 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 delivery/scavenging device was continuously used heart and dynamic image collection commenced as throughout the experiment. Once anesthesia was estab- described above, beginning approximately 30 sec prior to F-FDG, first over the heart lished, further isoflurane was carried by either air (20.9% injection of a second bolus of oxygen) or carbogen (95% oxygen/5% CO ). In order to and then over the tumor. All the studies described above inject F-FDG intravenously (IV), a cannula was inserted were performed while the mouse's body temperature was into a tail vein. The IV lines consisted of a 30 gauge 1/2" kept at 37°C. needle inserted into one end of a 60 cm length of PE10 tubing (IntraMedic Polyethylene) (internal diameter = An additional 2 mice were studied using this same two 0.28 mm; external diameter = 0.61 mm) and the needle of injection methodology but while their body temperature another removed from the hub inserted into the opposite was kept at 30°C. We have previously shown that C3H end of the tubing. IV lines were filled with heparin (100 SCCVII tumor-bearing animals placed on isoflurane expe- USP units/mL) prior to cannulation. To verify the integrity rience an approximate 7°C decrease in core body temper- of the line, a bolus of heparin was given prior to F-FDG ature 15 min after initiation of isoflurane [27]. To conduct injection. A bladder catheter (PE10 tubing) was used to low temperature studies animals were prepared as collect urine output and a rectal temperature probe (FISO described above and were administered isoflurane with- Technologies Fiber Optic Temperature Gauge Model FOT out temperature control until the core body temperature C-PEEK) was used to monitor and maintain core body dropped to 30°C at which point the warm air circulator temperature. Once these procedures were completed the was adjusted to maintain a core body temperature of mouse was immobilized in a jig that secured the feet. This 30°C. This temperature was maintained throughout the jig was inserted into a cylindrical Lucite chamber, fixed two injection study. with a port for the attachment of thermostat-controlled, warm air circulator (Nikon Model ITC-32, Nikon Inc. Blood sampling Japan) in order to maintain the mouse's core temperature. After anesthetizing the mouse with isoflurane carried by The entire assembly was fixed to a computer-controlled medical air (1.5%, 700 mL/min), the skin and integument motorized gantry capable of precisely moving the animal over the right or left ventral surface of the neck was cut into the scanner. After core temperature of 37°C was with surgical scissors and the cleidomastoideus and ster- established, PET image collection was started, and a 100 nomastoideus muscles were dissected to expose the left μl bolus of F-FDG (~400 μCi in 0.9% saline) was external jugular vein, removing fat and/or cauterizing injected. The injected doses were recorded for each animal peripheral vessels to minimize blood loss. A cannula, like imaged. Initial pilot studies were performed on 4 mice by that described previously for tail vein cannulation, was centering the tumor in the field-of-view prior to injection. then inserted into the external jugular vein and fixed in Two of these first mice were studied only with carbogen place with Vetbond tissue adhesive. A junction was made and two only with air. For each mouse, a dynamic acqui- by soldering a 30 G 1/2" needle into a 23 G 1" needle. The sition of 5 min/frame was acquired for 90 min total. In 30 G 1/2" needle end could be inserted into the end of the subsequent experiments each mouse was studied sequen- central line, while the 23 G 1" needle end could then be tially on both air and carbogen. These sequential studies, inserted into a small piece of PE50 tubing, into which a 10 performed on an additional 5 mice were more complex, μL Hamilton syringe could be inserted for a more precise each consisting of 5 phases. Initially, each animal was collection of blood. After administering F-FDG (~400 μCi, 100 μl) bolus, 10 μL of blood were drawn from the anesthetized using either air or carbogen as the carrier gas for the isoflurane (3 with air and 2 with carbogen). While central line every ten minutes for one hour. Each blood this gas was continuously administered, the scanner was sample was then diluted in 1 mL of saline, and its activity positioned at the level of the heart, and dynamic imaging was read in a Cobra II AutoGamma scintillation counter. begun immediately prior to injection of F-FDG (12 frames/10 sec for 2 min, 6 frames/30 sec, followed by 45 Image-derived and blood-draw derived input functions frames/1 min), continuing for 50 min, in order to capture It is possible that the input function (the concentration of an image-derived input function. The scanner was then arterial F-FDG as a function of time) would change moved to the level of the tumor (the lower extremities), depending on whether the mouse was breathing air or car- and dynamic imaging (5 min/frame) resumed for an addi- bogen. The area under the input function is a measure of tional 40 min, while continuing to administer the same the amount of F-FDG available for the tumor to metab- mixture of isoflurane and either air or carbogen. At the olize. In order to compare air and carbogen F-FDG end of this acquisition, the carrier gas for isoflurane was uptakes, the ratio of mean air to carbogen uptake in the switched from air to carbogen, or vice versa, and the ani- tumor volume of interest (VOI) was computed, normaliz- mal was left to equilibrate on the new gas mixture for 10 ing by the integral of the input function from time zero up min. When the 10-minute equilibration period had to the time of the measurement of uptake. The input func- ended, the scanner was again moved to the level of the Page 7 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 tion was estimated and used for normalization as air air air air ai ir described below. aA*(t) At () At () −b *B () t t t m == equation 2 carbo carbo carbo carbo carbo aA*(t) At () At () −b *B ()t t t m PET images at early time points revealed a large vessel dor- sal and/or caudal to the heart. Comparison with CT Thus, we only needed to correct the input function for images indicated that the vessel was the inferior vena cava, background contamination. Four venous blood samples but this could not be determined with absolute certainty were drawn from 30–60 min post injection. These sam- due to possible small mis-registrations between PET and ples were thought to be at late enough times so that CT. A strict determination of whether the vessel was arte- venous activity concentrations would be nearly identical rial or venous was thought unnecessary, apart from any to arterial concentrations, so A (t) = Blood(t) at late times. first transit uptake by the lung (known to be low), espe- This was confirmed by the very slow variation of A (t) cially since the bolus was slow compared to the expected versus t at these times. Therefore, for each dataset, air or rapid cardiac transit times [32]. carbogen, the scaling factor b was obtained considering the true blood activity derived from the blood draws at (t), To determine a measured blood time activity curve, A late time, the actual measured input function and the using the major blood vessel, regions of interest were background activity: drawn around this vessel at several transaxial levels to determine the relative mean activity concentration in the t 4 A () t − Blood() t vessel over time. Only relative concentrations could be b = equation 3 4 Bt () determined because the diameter of this vessel was tt = 1 thought to be small compared to the spatial resolution of To make the correction more robust, 4 time points were the scanner, making partial volume effects important. Fur- considered in order to estimate b. thermore, tissues adjacent to the vessel could have poten- tially contaminated the activity measured in the region of For logistical reasons, blood samples could not be drawn interest [33]. Corrections for these effects were made in from the same mouse. Therefore a cohort of 6 mice of the the following way. A square region of interest (ROI), 2.8 same age, size, and tumor development as the mice mm on a side and encompassing the aorta was drawn and imaged were injected with the same volume of FDG, and used to generate an aortic time activity curve. Two concen- each breathed either air or carbogen under the same con- tric rectangles were then drawn around this first ROI (7.3 ditions as the imaged mice. The time activity curves for all mm and 9.6 mm on a side) and a background region was mice (n = 3, air; n = 3, carbogen) were normalized by the obtained by subtracting the smallest ROI from the largest 18 activities of injected F-FDG and the curves were fit tem- ROI. Another time activity curve was created from this porally with a bi-exponential function. These fitted nor- background region in order to estimate the background malized blood samples were then used in the equation 3 activity, B(t). It was assumed that the measured input to compute b. Once b was determined for each imaged function, A (t) was equal to some combination of the m mouse, the true arterial ratio between air and carbogen true input function, A (t), and the background activity, t could be computed from equation 2. B(t): Volumes of interest (t) = a* A (t) + b*B(t). equation 1 m t Volumes of interest around the whole tumor were defined using a semi-automatic, three-dimensional, threshold The constant "a" is less than unity because some of the based, region-growing algorithm (MedX, Sensor Systems counts blur out of the vessel region of interest due to the Inc). The threshold was based on a percentage of the aver- partial volume effect. Similarly "b" represents blurring of age maximum intensity in the tumor. Average maximum background counts into the vessel's region of interest. B(t) intensity was defined as the maximum pixel in the tumor was only a negligibly small fraction of A(t) until late times averaged with neighboring pixels in a 0.5625 mm radius (> 4 min) so the correction was only important at late sphere, to reduce statistical noise. A 30% threshold gener- times. ated a VOI that visually included the entire tumor, but VOIs using other thresholds were also studied. Since we were only interested in estimating the ratio of the blood input functions as a result of two different physio- The very small in-plane mis-registrations of certain mice logical challenges, air and carbogen breathing, we did not that occurred between the air and the carbogen measure- need to correct the input functions for "a" because the size ments were corrected by maximizing the 3D correlation and placement of the ROI was identical in each case: between the two image sets (FLIRT linear registration tool; Image Analysis group, Oxford University). The final series of images collected during carbogen breathing were Page 8 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 in advanced cancer of the uterine cervix. Cancer Res 1996, aligned to the respective final series of images collected 56:4509-4515. during air breathing. Image data outside the tumor were 14. Griffiths JR, Taylor NJ, Howe FA, Saunders MI, Robinson SP, Hoskin not used for registration. The 3D-to-3D registration was PJ, Powell ME, Thoumine M, Caine LA, Baddeley H: The response of human tumors to carbogen breathing, monitored by gra- performed using six degrees of freedom (rigid body), with dient-recalled echo magnetic resonance imaging. Int J Radiat 3 translations and 3 rotations. After proper normalization Oncol Biol Phys 1997, 39:697-701. 15. Taylor NJ, Baddeley H, Goodchild KA, Powell ME, Thoumine M, Cul- for input function, the air data set was then subtracted ver LA, Stirling JJ, Saunders MI, Hoskin PJ, Phillips H, Padhani AR, Grif- from the carbogen data set, and the alignment of tumor fiths JR: BOLD MRI of human tumor oxygenation during signals was judged by eye. Although not used for align- carbogen breathing. J Magn Reson Imaging 2001, 14:156-163. 16. Clavo AC, Brown RS, Wahl RL: Fluorodeoxyglucose uptake in ment, the femoral vessels also appeared to be present in human cancer cell lines is increased by hypoxia. J Nucl Med each data set, and the signal in those regions was zero after 1995, 36:1625-1632. the subtraction, lending support to accurate registration. 17. Clavo AC, Wahl RL: Effects of hypoxia on the uptake of triti- ated thymidine, L-leucine, L-methionine and FDG in cul- Registration resulted in at most a 2 mm total displacement tured cancer cells. J Nucl Med 1996, 37:502-506. in the region of the tumor. 18. Pugachev A, Ruan S, Carlin S, Larson SM, Campa J, Ling CC, Humm JL: Dependence of FDG uptake on tumor microenvironment. Int J Radiat Oncol Biol Phys 2005, 62:545-553. Authors' contributions 19. Kahn BB, Flier JS: Regulation of glucose-transporter gene LWC, SE, JS, ALS, MVG, and JBM carried out animal imag- expression in vitro and in vivo. Diabetes Care 1990, 13:548-564. 20. Burgman P, Odonoghue JA, Humm JL, Ling CC: Hypoxia-Induced ing studies. SH, JC, and SLB conducted image analysis/ increase in FDG uptake in MCF7 cells. J Nucl Med 2001, modeling studies. LWC, JBM, MCK, and SLB drafted the 42:170-175. 21. Adam MF, Dorie MJ, Brown JM: Oxygen tension measurements manuscript. JBM conceived the study, and JBM, MVG, of tumors growing in mice. Int J Radiat Oncol Biol Phys 1999, SLB, MCK, and LWC participated in the design of the 45:171-180. study. 22. Powell ME, Collingridge DR, Saunders MI, Hoskin PJ, Hill SA, Chaplin DJ: Improvement in human tumour oxygenation with carbo- gen of varying carbon dioxide concentrations. Radiother Oncol Acknowledgements 1999, 50:167-171. This research was supported in part by the Intramural Research Program 23. Aquino-Parsons C, Lim P, Green A, Minchinton AI: Carbogen inha- lation in cervical cancer: assessment of oxygenation change. of the NIH, National Cancer Institute, Center for Cancer Research. Gynecol Oncol 1999, 74:259-264. 24. Kaanders JH, Bussink J, van der Kogel AJ: ARCON: a novel biol- References ogy-based approach in radiotherapy. Lancet Oncol 2002, 1. Chapman JD, Bradley JD, Eary JF, Haubner R, Larson SM, Michalski JM, 3:728-737. Okunieff PG, Strauss HW, Ung YC, Welch MJ: Molecular (func- 25. Bentzen L, Keiding S, Horsman MR, Falborg L, Hansen SB, Overgaard tional) imaging for radiotherapy applications: an RTOG sym- J: Feasibility of detecting hypoxia in experimental mouse posium. Int J Radiat Oncol Biol Phys 2003, 55:294-301. tumours with 18F-fluorinated tracers and positron emission 2. Coleman RE: FDG imaging. Nucl Med Biol 2000, 27:689-690. tomography – a study evaluating [18F]Fluoro-2-deoxy-D- 3. Teicher BA, Lazo JS, Sartorelli AC: Classification of antineoplas- glucose. Acta Oncol 2000, 39:629-637. tic agents by their selective toxicities toward oxygenated 26. Minn H, Clavo AC, Wahl RL: Influence of hypoxia on tracer and hypoxic tumor cells. Cancer Res 1981, 41:73-81. accumulation in squamous-cell carcinoma: in vitro evalua- 4. Moulder JE, Rockwell S: Hypoxic fractions of solid tumors: tion for PET imaging. Nucl Med Biol 1996, 23:941-946. experimental techniques, methods of analysis, and a survey 27. Reijnders K, English SJ, Krishna MC, Cook JA, Sowers AL, Mitchell JB, of existing data. Int J Radiat Oncol Biol Phys 1984, 10:695-712. Zhang Y: Influence of body temperature on the BOLD effect 5. Overgaard J, Horsman MR: Modification of hypoxia-induced in murine SCC tumors. Magn Reson Med 2004, 51:389-393. radioresistance in tumors by the use of oxygen and sensitiz- 28. Seidel J, Vaquero J, Green MV: Resolution uniformity and sensi- ers. Sem Radiat Oncol 1996, 6:10-121. tivity of the NIH ATLAS small animal PET scanner. Nucl Sci 6. Brown JM: Tumor microenvironment and the response to Symp Conf Rec, IEEE 2001, 3:1555-1558. anticancer therapy. Cancer Biol Ther 2002, 1:453-458. 29. Seidel J, Vaquero J, Pascau J, Desco M, Johnson CA, Green MV: Fea- 7. Khalil AA, Horsman MR, Overgaard J: The importance of deter- tures of the NIH ATLAS small animal PET scanner and its mining necrotic fraction when studying the effect of tumour use with a coaxial small animal volume CT scanner. Biomed volume on tissue oxygenation. Acta Oncol 1995, 34:297-300. Imag Biol 2002, 4:545-548. 8. Krishna MC, English S, Yamada K, Yoo J, Murugesan R, Devasahayam 30. Seidel J, Vaquero JJ, Green MV: Resolution uniformity and sensi- N, Cook JA, Golman K, Ardenkjaer-Larsen JH, Subramanian S, Mitch- tivity of the NIH ATLAS small animal PET scanner: compar- ell JB: Overhauser enhanced magnetic resonance imaging for ison to simulated LSO scanners without depth-of- tumor oximetry: coregistration of tumor anatomy and tis- interaction capability. IEEE Transactions on Nuclear Science 2003, sue oxygen concentration. Proc Natl Acad Sci 2002, 99:2216-2221. 50:1347-1350. 9. Halpern HJ, Yu C, Peric M, Barth E, Grdina DJ, Teicher BA: Oxyme- 31. Johnson CA, Seidel J, Carson RE, Gandler WR, Sofer A, Green MV, try deep in tissues with low-frequency electron paramag- Daube-Witherspoon ME: Evaluation of 3D reconstruction algo- netic resonance. Proc Natl Acad Sci USA 1994, 91:13047-13051. rithms for a small animal PET camera. IEEE Transactions on 10. Velan SS, Spencer RG, Zweier JL, Kuppusamy P: Electron paramag- Nuclear Science 1997, 44:1303-1308. netic resonance oxygen mapping (EPROM): direct visualiza- 32. Krivitski NM, Starostin D, Smith TL: Extracorporeal recording of tion of oxygen concentration in tissue. Magn Reson Med 2000, mouse hemodynamic parameters by ultrasound velocity 43:804-809. dilution. ASAIO J 1999, 45:3-36. 11. Warburg OH: The Metabolism of Tumors London: Constable Publica- 33. Watabe H, Channing MA, Riddell C, Jousse F, Libutti SK, Carrasquillo tions; 1930. JA, Bacharach SL, Carson RE: Noninvasive estimation of the 12. Bentzen L, Keiding S, Horsman MR, Gronroos T, Hansen SB, Over- aorta input function for measurement of tumor blood flow gaard J: Assessment of hypoxia in experimental mice tumours with [O-15]water. IEEE Trans Med Imag 2001, 20:164-174. by [18F]fluoromisonidazole PET and pO2 electrode meas- urements. Influence of tumour volume and carbogen breathing. Acta Oncol 2002, 41:304-312. 13. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P: Asso- ciation between tumor hypoxia and malignant progression Page 9 of 9 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Radiation Oncology Springer Journals

The influence of tumor oxygenation on 18F-FDG (Fluorine-18 Deoxyglucose) uptake: A mouse study using positron emission tomography (PET)

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/the-influence-of-tumor-oxygenation-on-18f-fdg-fluorine-18-deoxyglucose-aL0ul3DCPL

References (43)

Publisher
Springer Journals
Copyright
Copyright © 2006 by Chan et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; Oncology; Radiotherapy
eISSN
1748-717X
DOI
10.1186/1748-717X-1-3
pmid
16722588
Publisher site
See Article on Publisher Site

Abstract

Background: This study investigated whether changing a tumor's oxygenation would alter tumor metabolism, and thus uptake of F-FDG (fluorine-18 deoxyglucose), a marker for glucose metabolism using positron emission tomography (PET). Results: Tumor-bearing mice (squamous cell carcinoma) maintained at 37°C were studied while breathing either normal air or carbogen (95% O , 5% CO ), known to significantly oxygenate 2 2 tumors. Tumor activity was measured within an automatically determined volume of interest (VOI). Activity was corrected for the arterial input function as estimated from image and blood- derived data. Tumor FDG uptake was initially evaluated for tumor-bearing animals breathing only air (2 animals) or only carbogen (2 animals). Subsequently, 5 animals were studied using two sequential F-FDG injections administered to the same tumor-bearing mouse, 60 min apart; the first injection on one gas (air or carbogen) and the second on the other gas. When examining the entire tumor VOI, there was no significant difference of F-FDG uptake between mice breathing either air or carbogen (i.e. air/carbogen ratio near unity). However, when only the highest F-FDG uptake regions of the tumor were considered (small VOIs), there was a modest (21%), but significant increase in the air/carbogen ratio suggesting that in these potentially most hypoxic regions of the tumor, F-FDG uptake and hence glucose metabolism, may be reduced by increasing tumor oxygenation. Conclusion: Tumor F-FDG uptake may be reduced by increases in tumor oxygenation and thus may provide a means to further enhance F-FDG functional imaging. Page 1 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 through the tumor of a mouse (maintained at 37°C) Background F-FDG (fluorine-18 deoxyglucose) has become widely breathing air and a mouse breathing carbogen. Figure 1, used as a radiolabeled marker for positron emission tom- bottom, shows the time activity curve obtained from the ography (PET) imaging of solid tumors. In some regions VOIs drawn around the entire tumor for a tumor-bearing of the body, the predictive power for identifying cancer animal breathing air. The ordinate of the time activity 18 18 using F-FDG approaches 95% [1,2]. curve is the relative F-FDG activity concentration in the tumor per mCi injected dose. Note that the decay cor- While altered glucose metabolism is a unique feature of rected FDG uptake curve is quite flat from 60 min until the neoplastic growth, there are other factors associated with end of the study (130 min). This was true for all four mice the tumor micro-environment that are in marked contrast in the pilot study. Similar time activity curves were to normal tissues. The vascular architecture in tumor tis- obtained for animals breathing carbogen (data not sue is abnormal and differs greatly from normal tissues, shown). Note also that the uptake in these two mice (one resulting in altered blood flow and the development of on air and one on carbogen) was visually similar with tumor hypoxia. The presence of tumor hypoxia is thought considerable heterogeneity of F-FDG uptake across the to represent a barrier for effective cancer treatment for tumor. Air versus carbogen comparisons could not be both radiation and chemotherapy [3-6]. Identifying made for these animals however, because data for the patients whose tumors contain hypoxic areas may there- mice in the pilot studies could not be corrected for possi- fore have an important role in tumor prognosis as well as ble differences in the arterial input function. In addition, treatment approach and outcome. Currently, the ability to F-FDG uptake would presumably have mouse-to-mouse identify and quantify tumor hypoxia is limited. The cur- physiologic variation since different mice were used for air rent gold standard for measuring tissue oxygen concentra- and for carbogen, potentially masking true differences in tion utilizes oxygen-sensitive electrodes, which measure uptake caused by the nature of the ventilating gas. For oxygen partial pressure (pO ) directly in tumor tissue. Given the invasive nature of this technique, it is difficult to access deep-seated tumors, and once assessed, it is dif- ficult to distinguish between measurements made in necrotic and viable regions [7]. Non-invasive techniques, such as Overhauser-enhanced magnetic resonance imag- ing (OMRI) [8] and electron paramagnetic resonance imaging (EPRI) [9,10], are being evaluated to avoid the obstacles encountered with the polarographic electrode. Since non-invasive F-FDG/PET imaging is already widely used in clinical facilities to identify malignant tis- sue, we questioned whether this imaging modality might be used to assess tumor hypoxia. Anaerobic metabolism requires much more glucose to generate the same amount of ATP as under aerobic metabolism. Given the hypoxic nature of certain tumors, regions of tumor tissue are known to resort to anaerobic glucose metabolism [11]. We hypothesize that experimentally increasing a tumor's oxygen supply might enable the tumor to metabolize more glucose aerobically, thus reducing F-FDG uptake. To test this hypothesis, we have compared tumor F-FDG uptake in tumor-bearing animals sequentially breathing air (20.9% O ) and carbogen (95% oxygen, 5% CO ). 2 2 ial slices through the breathing a Figure 1 F-FDG images ( ir and a differ top) showin tumor of ent mouse breathing ca g sa a mouse (maintained gittal, coronal, an rbogen d transax- at 37°C) Carbogen breathing has been shown to markedly increase F-FDG images (top) showing sagittal, coronal, and transax- the oxygenation status of tumors [8,12]. In addition, we ial slices through the tumor of a mouse (maintained at 37°C) present data stressing the importance of maintaining nor- breathing air and a different mouse breathing carbogen. Both mal body temperature in small animals undergoing F- color scales set to the same maximum nCi/cc/injected-dose. FDG/PET imaging. Bottom: time activity curve obtained from the VOIs drawn around the entire tumor for a tumor-bearing animal breath- Results ing air (animals breathing carbogen also resulted in similar curves, flat at late times). Figure 1 shows images from two of the pilot studies. Fig- ure 1 (top) shows sagittal, coronal, and transaxial slices Page 2 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 these reasons the second more elaborate set of studies described above were undertaken. This second set of stud- F-FDG injections in the ies involved two sequential same mouse, 60 minutes apart, the first injection on one gas (air or carbogen) and the second on the other gas. The fact that the F-FDG time activity curves from the pilot studies were flat, regardless of whether air or carbogen was used, meant that in the two injection studies, we could subtract the F-FDG activity present at the end of first 60 min from the activity following the second injection. These two injection studies permitted sequential measure- ment of air and carbogen uptake in the same tumor, and Blood- and normal physiologic cor Figure 2 image-derived input functions fo e body temperature (37° r mice sca C) nned at permitted correction for possible differences in the arterial Blood- and image-derived input functions for mice scanned at input function. normal physiologic core body temperature (37°C). Blood- derived input functions were determined using blood sam- Air to carbogen uptake ratios at normal temperature 18 ples drawn every 10 min, starting 30 min after FDG injec- (37°C) tion. Thus, the blood-derived input function curve begins at time = 30 min. Note that the blood samples always yielded For each mouse we computed the ratio of the uptake lower activity concentration than the image data due to while breathing air to the uptake while breathing carbo- background activity, as per equation 1. gen, each corrected for the area under the input function as per equation 2. A typical input function is shown in Fig- ure 2, along with the late blood sample derived input function. The measured input function was corrected even if the maximum tumor value (averaged with its clos- using the blood samples, as per equations 1–3. The ratio est neighboring pixels in all 3 dimensions) was used for of corrected tumor uptake while breathing air to corrected each tumor. We wished to visualize which pixels were tumor uptake while breathing carbogen, averaged over all causing the difference in F-FDG tumor uptake for air ver- sus carbogen breathing. Figure 4 shows F-FDG tumor 5 mice, did not differ significantly from 1.0 (ratio air/car- bogen = 1.15, p = NS, Figure 3, squares). The ratio uptake images for three different animals who received remained not significantly different from unity whether two F-FDG injections, one while breathing air and the the VOI was drawn using a 30, 40 or 50% threshold, and other while breathing carbogen. The first column shows even when manually drawn VOIs encompassing the entire tumor were used. Note from Figure 3 that the tumors of four out of five mice exhibited air/carbogen ratios, all >1, while one mouse deviated from the other 4, with a ratio lower than unity. There was no a priori reason to exclude the outlying animal, but doing so resulted in an air/carbo- gen ratio, which did differ significantly from unity (air/ carbogen = 1.22, p < 0.01). As shown in Figure 1 the SCC tumor exhibited considera- ble heterogeneity in its uptake of F-FDG. We hypothe- sized that regions of the tumor demonstrating accentuated uptake of FDG might be exceptionally hypoxic and so might respond differently to an increase in the concentration of inspired oxygen as compared to other tumor tissue. To test this hypothesis we analyzed small VOI's (90% threshold) around regions of the Air/ca VOI (squar progra the tumor (9 Figure 3 rbogen ratios of m using a es), as determined with the 3D region 0% threshold) 30% threshold, and for tumor FDG uptake fo the hottest r whol growing part of e tumor tumors demonstrating the greatest activity. Unlike the Air/carbogen ratios of tumor FDG uptake for whole tumor data from the whole tumor, the uptake (corrected for VOI (squares), as determined with the 3D region growing input function) for these small hot VOIs resulted in an air/ program using a 30% threshold, and for the hottest part of carbogen ratio that was consistent across all 5 mice, and the tumor (90% threshold). Tumor uptake ratios were cor- was significantly greater than unity (air/carbogen = 1.21, rected using derived input function data as per equation 2 nd p < 0.005, Figure 3, right). Again this result was insensitive and the 2 injection activity was corrected for residual activ- ity remaining as a result of the first injection. to the way the VOI was defined. The results held when the VOI was based on either an 80% or 90% threshold, or Page 3 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 ing point in Figure 3. Negative values are not displayed in column 3. Instead the fourth column shows the (many fewer) pixels in which carbogen exceeded air (i.e. the neg- ative of the equation used for column 3). Air to carbogen uptake ratios at low temperature (30°C) When the core temperature of the mice was maintained at 30°C, an air/carbogen tumor ratio (averaged over all mice) very much less than one was observed (mean air/ carbogen ratio = 0.47 ± 0.30; n = 2). Again, this result was insensitive to the manner of VOI definition. Figure 5 shows F-FDG uptake during carbogen and air for a low temperature mouse. It is quite clear that the tumor uptake while breathing carbogen is markedly higher than the uptake while breathing air. However, whereas the blood- 18 18 S receiving two injections of b carbogen breath 3 Figure 4 agittal r 7eathing °C throughout the study) FDG tumo air and the ing, or vice-v r uptak second afte e FDG, ers images r the a ( one whi ani fo animal was switched to mals r three mice (A, lmaintaine e the animal w d at B, C) as derived input functions from mice at normal physiologic Sagittal FDG tumor uptake images for three mice (A, B, C) temperature (37°C) were highly reproducible across mul- receiving two injections of FDG, one while the animal was tiple mice, there was a large variability between blood- breathing air and the second after the animal was switched to derived input functions in the low-temperature mice. carbogen breathing, or vice-versa (animals maintained at Thus, one must interpret the data from mice not main- 37°C throughout the study). The first column shows the rel- ative tumor F-FDG uptake for air breathing. The second tained at physiologic temperature with caution. column represents tumor FDG uptake for the animal after switching to carbogen breathing (residual F-FDG from ini- Discussion tial injection subtracted). The first and second columns are Hypoxia in human tumors can significantly influence shown with the same relative color scale. The third column treatment outcome and the aggressiveness of the tumor shows the pixels in which air uptake exceeded carbogen [13]. This finding has led to considerable interest in non- uptake, expressed as 100*(AirImage - CarbogenImage)/ invasive means to assess the extent of hypoxia in tumors (mean counts in Air tumor VOI). Ideally the denominator prior to therapy, for example with MR (BOLD) imaging would have been the air image, but due to image noise con- [14,15]. Several studies have shown that tumor cell or siderations the normalization described was done instead. tumor F-FDG uptake is associated with hypoxia [16-18]. Negative values are not displayed in column 3. Instead the fourth column shows the (many fewer) pixels in which car- Glucose transporter (GLUT) receptor proteins and hexoki- bogen exceeded air (i.e. the negative of the equation used for nase activity are elevated in tumors and thought to be column 3. Corrections were made for injected activity and responsible for increased F-FDG tumor uptake com- for residual activity. The numeric values shown on the color pared to normal tissues [19]. Hypoxia has been shown to bar apply only to columns 3 and 4. the tumor F-FDG uptake for air breathing. The second column represents tumor F-FDG uptake for the animal after switching to carbogen breathing (residual F-FDG from initial injection subtracted). From these images it is difficult to tell which pixels were causing the slight decrease in tumor F-FDG uptake for animals breathing carbogen. To clarify this, the third column shows the per- centage increase in F-FDG uptake for air breathing ani- mals versus carbogen breathing animals expressed as 18 18 and the second while Figure 5 slices through th given F-FDG images showing sagittal, c two injections of e tumor breathing carbogen F-FDG, the first wh of a mouse ma oronal, inta and transaxial ined at 30 ile breathing air °C 100* (AirImage – CarbogenImage)/(mean counts in Air F-FDG images showing sagittal, coronal, and transaxial tumor VOI). Ideally the denominator would have been slices through the tumor of a mouse maintained at 30°C the air image, but due to image noise considerations the 18 given two injections of F-FDG, the first while breathing air mean value in the air tumor VOI (30% threshold) were and the second while breathing carbogen. Both images are used instead. The data were normalized so that when the displayed to the same scale, and are corrected for injected activity and residual activity produced by the first injection. tumor VOI was placed on the percentage increase image, it gave a value equal to the value shown in the correspond- Page 4 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 be a significant factor in the over-expression of these pro- carbogen ratios for overall tumor F-FDG uptake were teins [16,17,20]. In the present study we examined not significantly different from unity. However, when F-FDG uptake region of each tumor was whether information concerning hypoxia could be gained only the highest using F-FDG and PET imaging. This was considered pos- considered (perhaps the most hypoxic regions) the hypo- sible since the hypoxic nature of most tumors forces thesis was supported. In these presumably metabolically tumor glucose metabolism to utilize primarily glycolytic active regions, there was a small (21%) but significant pathways, rather than oxidative metabolism [11]. Very increase in air/carbogen uptake, suggesting reduced F- large quantities of glucose are required to produce a given FDG uptake under conditions of better oxygenation. The amount of ATP via the glycolytic pathway compared to lack of an air/carbogen difference for the whole tumor, the oxidative pathway, which supposedly accounts for and the fairly small air/carbogen difference even for the why many tumors have such high F-FDG uptake. We hottest part of the tumor might suggest that tumors are hypothesized that by oxygenating the tumor (via breath- "programmed" to burn glucose primarily in an anaerobic ing carbogen), tumor glucose metabolism would begin to fashion, regardless of their level of oxygenation. This have a significant oxidative component, requiring consid- could be a potential survival mechanism for tissue des- erably less glucose per ATP molecule produced. Thus, bet- tined to grow in an anaerobic environment. This potential ter oxygenated tumors (i.e. tumors influenced by explanation is of course speculative. Another factor which breathing carbogen) should have less F-FDG uptake could play a role is the potential vasodilatory properties of than when in their usual, hypoxic state (i.e. while the ani- carbon dioxide, although there is conflicting data mal is breathing air). To examine this possibility, F-FDG reported in the literature [24]. uptake in tumors was examined in tumors when the mouse was breathing air versus carbogen. Changes in tumor FDG uptake (hot spots) when breath- ing carbogen might add another dimension to F-FDG The SCC tumor used in the present study is known to be functional imaging. While small, the 21% change in F- quite hypoxic as determined by previous oxygen electrode FDG uptake in small regions may be giving information measurements in our laboratory [8]. Even though the similar and perhaps complimentary to that obtained by tumor is extremely hypoxic, there was no evidence of the MRI based oxygen imaging is Blood Oxygen Level necrosis in the tumors used in the present study (~1 cm Dependent (BOLD) MRI. For this imaging technique, diameter). Carbogen breathing has been shown in numer- temporal changes in the ratios of deoxy- to oxyhemo- ous studies to increase the oxygenation status of tumor- globin can be monitored utilizing the contrast mecha- bearing mice [8,21] and in patients [22-24]. Using the nisms provided by deoxyhemoglobin to protons. With same SCCVII tumor model, tumor size, and experimental this method, it is possible to examine changes in blood conditions with respect to the maintenance to core body flow and tumor oxygenation levels in response to carbo- temperature as used in the present study, we have previ- gen breathing versus air breathing [14,15]. The enhance- ously shown that carbogen breathing increases oxygena- ment observed in BOLD MRI images is typically only a tion of the SCCVII tumor as measured by Eppendorf few percent. In contrast, the decrease in F-FDG tumor oxygen electrodes [8,8]. In this study the mean tumor pO uptake upon carbogen breathing in small regions of the values for animals breathing air was 8.2 mm Hg, which tumor in the present study was up to 21%, suggesting that increased to a value of 19.8 for animals breathing carbo- in these small regions tumor glucose metabolism is gen [8]. altered as a result of increased perfusion and oxygenation. Should this observation be correct, which will require fur- As mentioned above, the hypoxic nature of the SCC tumor ther validation, the added functional dimension to F- makes it likely that a large proportion of the tumor cells FDG imaging afforded by carbogen breathing could be would utilize glucose by anaerobic glycolysis, and so useful to clinicians assessing treatment response, particu- exhibit enhanced F-FDG uptake. Indeed, we found glu- larly for radiation treatment. cose utilization as measured by uptake of F-FDG to be significant in the SCC tumor as evidenced by the tumor Our results did not wholly agree with some previous stud- images shown in Figure 1. The FDG uptake in the tumor ies that showed whole tumor F-FDG uptake increases in was particularly striking compared to the non-tumor bear- more hypoxic conditions [25-28]. Our whole tumor ing leg where little to no uptake was observed (data not results gave an air/carbogen uptake ratio that did not dif- shown). It was anticipated that increasing the oxygenation fer significantly from unity. Only when VOIs were drawn of the tumor by carbogen breathing would lead to a shift around the hottest part of our tumors were our findings to aerobic metabolism producing much more ATP per more consistent with these previous studies. The afore- molecule of glucose, thereby decreasing the need for glu- mentioned studies either examined a different tumor type cose, and decreasing F-FDG tumor uptake. The results of from our study (C3H mammary carcinomas) [25], or this study only partially supported this hypothesis. Air/ examined the same tumor type (SCC), but in an in vitro Page 5 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 setting [26]. These differences alone could potentially at 37°C. When only the highest uptake regions of the account for the difference in our findings. Additionally, tumor were considered, however, there was a modest this result could be due to differences in experimental (21%) but significant increase in the air/carbogen ratio. method, or due to the variability in our own results. Pre- This suggests that in these potentially most hypoxic vious studies did not attempt to correct for potential dif- regions of the tumor, F-FDG uptake may be reduced by ferences in input function. The lack of standardization increases in tumor oxygenation and thus may provide a across animal PET studies, both in the means of determin- means to further enhance F-FDG functional imaging. As ing reference input functions, as well as lack of consensus a secondary finding, we note that when the mouse's tem- regarding simpler procedures such as controlling animal perature was permitted to fall to 30°C (as occurs naturally core temperature, might also explain this difference in during isoflurane anesthesia), significant alterations in result. Future experiments with a greater number of mice, F-FDG metabolism were observed, making it clear that using simultaneous blood sampling, would clarify the strict physiological controls are necessary when such results shown in Figure 3. experiments are performed. The ability to examine and quantify regions of greatest Methods change in F-FDG uptake, which is where a radiation Animals and tumors oncologist would presumably concentrate three-dimen- Female C3H mice, produced by the National Cancer Insti- sional conformal radiation dose painting, has significant tute Animal Production Area (Frederick, MD), were used implications for cancer diagnosis and treatment. Future for this study. The mice were 7–9 weeks of age at the time studies could further test this phenomenon by shifting to of experimentation and weighed between 20–30 grams. more hypoxic conditions, to see if F-FDG uptake All experiments were carried out under a protocol improves even more. approved by the National Cancer Institute Animal Care and Use Committee, and were in compliance with the As an adjunct to our study, we examined the results Guide for the Care and Use Of Laboratory Animal Resource, obtained when mice were allowed to drop core body tem- (1996) National Research Council. Tumors were grown in perature to 30°C. Past work in our laboratory has demon- the mice by a subcutaneous (s.c.) injection of a single-cell squamous cell carcinoma (SCCVII) strated that animal's core body temperature under suspension of 2 × 10 isoflurane anesthesia drops ten degrees in an average of 15 cells in the right hind leg. Tumors grew to a size of ~1.0 min [27]. We found that mice at this lower temperature cm diameter 7–10 days after injection. Tumor volumes showed significantly greater F-FDG accumulation in car- were determined prior to PET scanning, by measuring bogen than air environments. This could be due to relative three orthogonal diameters using calipers. For all imaging vasoconstriction in low temperature conditions, possibly studies ~1 cm diameter tumors were used. making F-FDG uptake flow limited in these circum- stances. Adding carbogen to this low-temperature state PET scanner may improve blood flow, because vasculature responds to PET images were obtained with the ATLAS (Advanced the high partial pressure of oxygen by vasodilation. Technology Laboratory Animal Scanner), a dedicated Although this reasoning is speculative, our low tempera- small animal PET scanner developed at NIH with an axial ture findings do point definitively to the necessity for field-of-view (FOV) of 2 cm, a transverse FOV of 6.8 cm strict control of the mice's physiology and the need for and an aperture of 8 cm [28-30]. standardization across animal PET studies. The image data acquired with the ATLAS PET scanner were One strength of the present study was the ability to meas- reconstructed using the three-dimensional ordered subset ure F-FDG uptake with air and carbogen breathing in expectation maximization (OSEM) reconstruction algo- the same tumor. Previous studies have compared uptake rithm, including a model of the system resolution [31]. across different animals. Differences in tumor heterogene- All reconstructions were performed on the NIH Biowulf ity across mice potentially confound these results. Our computer cluster, utilizing sixteen subsets and taking 10 study controlled for this confounding variable. Further iterations. Under these conditions, a spatial resolution of studies would benefit from testing multiple tumor types, 1.6 mm full-width-at-half-maximum (FWHM) was as well as human tumors. achieved. PET scan experimental set-up Conclusion The ratio of F-FDG uptake in SCC tumors while breath- The mice were fasted at least 3 hours prior to injection of ing air to that while breathing carbogen, did not differ sig- F-FDG. Each mouse was anesthetized with isoflurane nificantly from unity when the entire tumor was (induction; 2.0%, maintenance; 1.5%) carried by air at considered, and when the animal's temperature was kept 750 ml/min and delivered via a nosepiece. An anesthetic Page 6 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 delivery/scavenging device was continuously used heart and dynamic image collection commenced as throughout the experiment. Once anesthesia was estab- described above, beginning approximately 30 sec prior to F-FDG, first over the heart lished, further isoflurane was carried by either air (20.9% injection of a second bolus of oxygen) or carbogen (95% oxygen/5% CO ). In order to and then over the tumor. All the studies described above inject F-FDG intravenously (IV), a cannula was inserted were performed while the mouse's body temperature was into a tail vein. The IV lines consisted of a 30 gauge 1/2" kept at 37°C. needle inserted into one end of a 60 cm length of PE10 tubing (IntraMedic Polyethylene) (internal diameter = An additional 2 mice were studied using this same two 0.28 mm; external diameter = 0.61 mm) and the needle of injection methodology but while their body temperature another removed from the hub inserted into the opposite was kept at 30°C. We have previously shown that C3H end of the tubing. IV lines were filled with heparin (100 SCCVII tumor-bearing animals placed on isoflurane expe- USP units/mL) prior to cannulation. To verify the integrity rience an approximate 7°C decrease in core body temper- of the line, a bolus of heparin was given prior to F-FDG ature 15 min after initiation of isoflurane [27]. To conduct injection. A bladder catheter (PE10 tubing) was used to low temperature studies animals were prepared as collect urine output and a rectal temperature probe (FISO described above and were administered isoflurane with- Technologies Fiber Optic Temperature Gauge Model FOT out temperature control until the core body temperature C-PEEK) was used to monitor and maintain core body dropped to 30°C at which point the warm air circulator temperature. Once these procedures were completed the was adjusted to maintain a core body temperature of mouse was immobilized in a jig that secured the feet. This 30°C. This temperature was maintained throughout the jig was inserted into a cylindrical Lucite chamber, fixed two injection study. with a port for the attachment of thermostat-controlled, warm air circulator (Nikon Model ITC-32, Nikon Inc. Blood sampling Japan) in order to maintain the mouse's core temperature. After anesthetizing the mouse with isoflurane carried by The entire assembly was fixed to a computer-controlled medical air (1.5%, 700 mL/min), the skin and integument motorized gantry capable of precisely moving the animal over the right or left ventral surface of the neck was cut into the scanner. After core temperature of 37°C was with surgical scissors and the cleidomastoideus and ster- established, PET image collection was started, and a 100 nomastoideus muscles were dissected to expose the left μl bolus of F-FDG (~400 μCi in 0.9% saline) was external jugular vein, removing fat and/or cauterizing injected. The injected doses were recorded for each animal peripheral vessels to minimize blood loss. A cannula, like imaged. Initial pilot studies were performed on 4 mice by that described previously for tail vein cannulation, was centering the tumor in the field-of-view prior to injection. then inserted into the external jugular vein and fixed in Two of these first mice were studied only with carbogen place with Vetbond tissue adhesive. A junction was made and two only with air. For each mouse, a dynamic acqui- by soldering a 30 G 1/2" needle into a 23 G 1" needle. The sition of 5 min/frame was acquired for 90 min total. In 30 G 1/2" needle end could be inserted into the end of the subsequent experiments each mouse was studied sequen- central line, while the 23 G 1" needle end could then be tially on both air and carbogen. These sequential studies, inserted into a small piece of PE50 tubing, into which a 10 performed on an additional 5 mice were more complex, μL Hamilton syringe could be inserted for a more precise each consisting of 5 phases. Initially, each animal was collection of blood. After administering F-FDG (~400 μCi, 100 μl) bolus, 10 μL of blood were drawn from the anesthetized using either air or carbogen as the carrier gas for the isoflurane (3 with air and 2 with carbogen). While central line every ten minutes for one hour. Each blood this gas was continuously administered, the scanner was sample was then diluted in 1 mL of saline, and its activity positioned at the level of the heart, and dynamic imaging was read in a Cobra II AutoGamma scintillation counter. begun immediately prior to injection of F-FDG (12 frames/10 sec for 2 min, 6 frames/30 sec, followed by 45 Image-derived and blood-draw derived input functions frames/1 min), continuing for 50 min, in order to capture It is possible that the input function (the concentration of an image-derived input function. The scanner was then arterial F-FDG as a function of time) would change moved to the level of the tumor (the lower extremities), depending on whether the mouse was breathing air or car- and dynamic imaging (5 min/frame) resumed for an addi- bogen. The area under the input function is a measure of tional 40 min, while continuing to administer the same the amount of F-FDG available for the tumor to metab- mixture of isoflurane and either air or carbogen. At the olize. In order to compare air and carbogen F-FDG end of this acquisition, the carrier gas for isoflurane was uptakes, the ratio of mean air to carbogen uptake in the switched from air to carbogen, or vice versa, and the ani- tumor volume of interest (VOI) was computed, normaliz- mal was left to equilibrate on the new gas mixture for 10 ing by the integral of the input function from time zero up min. When the 10-minute equilibration period had to the time of the measurement of uptake. The input func- ended, the scanner was again moved to the level of the Page 7 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 tion was estimated and used for normalization as air air air air ai ir described below. aA*(t) At () At () −b *B () t t t m == equation 2 carbo carbo carbo carbo carbo aA*(t) At () At () −b *B ()t t t m PET images at early time points revealed a large vessel dor- sal and/or caudal to the heart. Comparison with CT Thus, we only needed to correct the input function for images indicated that the vessel was the inferior vena cava, background contamination. Four venous blood samples but this could not be determined with absolute certainty were drawn from 30–60 min post injection. These sam- due to possible small mis-registrations between PET and ples were thought to be at late enough times so that CT. A strict determination of whether the vessel was arte- venous activity concentrations would be nearly identical rial or venous was thought unnecessary, apart from any to arterial concentrations, so A (t) = Blood(t) at late times. first transit uptake by the lung (known to be low), espe- This was confirmed by the very slow variation of A (t) cially since the bolus was slow compared to the expected versus t at these times. Therefore, for each dataset, air or rapid cardiac transit times [32]. carbogen, the scaling factor b was obtained considering the true blood activity derived from the blood draws at (t), To determine a measured blood time activity curve, A late time, the actual measured input function and the using the major blood vessel, regions of interest were background activity: drawn around this vessel at several transaxial levels to determine the relative mean activity concentration in the t 4 A () t − Blood() t vessel over time. Only relative concentrations could be b = equation 3 4 Bt () determined because the diameter of this vessel was tt = 1 thought to be small compared to the spatial resolution of To make the correction more robust, 4 time points were the scanner, making partial volume effects important. Fur- considered in order to estimate b. thermore, tissues adjacent to the vessel could have poten- tially contaminated the activity measured in the region of For logistical reasons, blood samples could not be drawn interest [33]. Corrections for these effects were made in from the same mouse. Therefore a cohort of 6 mice of the the following way. A square region of interest (ROI), 2.8 same age, size, and tumor development as the mice mm on a side and encompassing the aorta was drawn and imaged were injected with the same volume of FDG, and used to generate an aortic time activity curve. Two concen- each breathed either air or carbogen under the same con- tric rectangles were then drawn around this first ROI (7.3 ditions as the imaged mice. The time activity curves for all mm and 9.6 mm on a side) and a background region was mice (n = 3, air; n = 3, carbogen) were normalized by the obtained by subtracting the smallest ROI from the largest 18 activities of injected F-FDG and the curves were fit tem- ROI. Another time activity curve was created from this porally with a bi-exponential function. These fitted nor- background region in order to estimate the background malized blood samples were then used in the equation 3 activity, B(t). It was assumed that the measured input to compute b. Once b was determined for each imaged function, A (t) was equal to some combination of the m mouse, the true arterial ratio between air and carbogen true input function, A (t), and the background activity, t could be computed from equation 2. B(t): Volumes of interest (t) = a* A (t) + b*B(t). equation 1 m t Volumes of interest around the whole tumor were defined using a semi-automatic, three-dimensional, threshold The constant "a" is less than unity because some of the based, region-growing algorithm (MedX, Sensor Systems counts blur out of the vessel region of interest due to the Inc). The threshold was based on a percentage of the aver- partial volume effect. Similarly "b" represents blurring of age maximum intensity in the tumor. Average maximum background counts into the vessel's region of interest. B(t) intensity was defined as the maximum pixel in the tumor was only a negligibly small fraction of A(t) until late times averaged with neighboring pixels in a 0.5625 mm radius (> 4 min) so the correction was only important at late sphere, to reduce statistical noise. A 30% threshold gener- times. ated a VOI that visually included the entire tumor, but VOIs using other thresholds were also studied. Since we were only interested in estimating the ratio of the blood input functions as a result of two different physio- The very small in-plane mis-registrations of certain mice logical challenges, air and carbogen breathing, we did not that occurred between the air and the carbogen measure- need to correct the input functions for "a" because the size ments were corrected by maximizing the 3D correlation and placement of the ROI was identical in each case: between the two image sets (FLIRT linear registration tool; Image Analysis group, Oxford University). The final series of images collected during carbogen breathing were Page 8 of 9 (page number not for citation purposes) Radiation Oncology 2006, 1:3 http://www.ro-journal.com/content/1/1/3 in advanced cancer of the uterine cervix. Cancer Res 1996, aligned to the respective final series of images collected 56:4509-4515. during air breathing. Image data outside the tumor were 14. Griffiths JR, Taylor NJ, Howe FA, Saunders MI, Robinson SP, Hoskin not used for registration. The 3D-to-3D registration was PJ, Powell ME, Thoumine M, Caine LA, Baddeley H: The response of human tumors to carbogen breathing, monitored by gra- performed using six degrees of freedom (rigid body), with dient-recalled echo magnetic resonance imaging. Int J Radiat 3 translations and 3 rotations. After proper normalization Oncol Biol Phys 1997, 39:697-701. 15. Taylor NJ, Baddeley H, Goodchild KA, Powell ME, Thoumine M, Cul- for input function, the air data set was then subtracted ver LA, Stirling JJ, Saunders MI, Hoskin PJ, Phillips H, Padhani AR, Grif- from the carbogen data set, and the alignment of tumor fiths JR: BOLD MRI of human tumor oxygenation during signals was judged by eye. Although not used for align- carbogen breathing. J Magn Reson Imaging 2001, 14:156-163. 16. Clavo AC, Brown RS, Wahl RL: Fluorodeoxyglucose uptake in ment, the femoral vessels also appeared to be present in human cancer cell lines is increased by hypoxia. J Nucl Med each data set, and the signal in those regions was zero after 1995, 36:1625-1632. the subtraction, lending support to accurate registration. 17. Clavo AC, Wahl RL: Effects of hypoxia on the uptake of triti- ated thymidine, L-leucine, L-methionine and FDG in cul- Registration resulted in at most a 2 mm total displacement tured cancer cells. J Nucl Med 1996, 37:502-506. in the region of the tumor. 18. Pugachev A, Ruan S, Carlin S, Larson SM, Campa J, Ling CC, Humm JL: Dependence of FDG uptake on tumor microenvironment. Int J Radiat Oncol Biol Phys 2005, 62:545-553. Authors' contributions 19. Kahn BB, Flier JS: Regulation of glucose-transporter gene LWC, SE, JS, ALS, MVG, and JBM carried out animal imag- expression in vitro and in vivo. Diabetes Care 1990, 13:548-564. 20. Burgman P, Odonoghue JA, Humm JL, Ling CC: Hypoxia-Induced ing studies. SH, JC, and SLB conducted image analysis/ increase in FDG uptake in MCF7 cells. J Nucl Med 2001, modeling studies. LWC, JBM, MCK, and SLB drafted the 42:170-175. 21. Adam MF, Dorie MJ, Brown JM: Oxygen tension measurements manuscript. JBM conceived the study, and JBM, MVG, of tumors growing in mice. Int J Radiat Oncol Biol Phys 1999, SLB, MCK, and LWC participated in the design of the 45:171-180. study. 22. Powell ME, Collingridge DR, Saunders MI, Hoskin PJ, Hill SA, Chaplin DJ: Improvement in human tumour oxygenation with carbo- gen of varying carbon dioxide concentrations. Radiother Oncol Acknowledgements 1999, 50:167-171. This research was supported in part by the Intramural Research Program 23. Aquino-Parsons C, Lim P, Green A, Minchinton AI: Carbogen inha- lation in cervical cancer: assessment of oxygenation change. of the NIH, National Cancer Institute, Center for Cancer Research. Gynecol Oncol 1999, 74:259-264. 24. Kaanders JH, Bussink J, van der Kogel AJ: ARCON: a novel biol- References ogy-based approach in radiotherapy. Lancet Oncol 2002, 1. Chapman JD, Bradley JD, Eary JF, Haubner R, Larson SM, Michalski JM, 3:728-737. Okunieff PG, Strauss HW, Ung YC, Welch MJ: Molecular (func- 25. Bentzen L, Keiding S, Horsman MR, Falborg L, Hansen SB, Overgaard tional) imaging for radiotherapy applications: an RTOG sym- J: Feasibility of detecting hypoxia in experimental mouse posium. Int J Radiat Oncol Biol Phys 2003, 55:294-301. tumours with 18F-fluorinated tracers and positron emission 2. Coleman RE: FDG imaging. Nucl Med Biol 2000, 27:689-690. tomography – a study evaluating [18F]Fluoro-2-deoxy-D- 3. Teicher BA, Lazo JS, Sartorelli AC: Classification of antineoplas- glucose. Acta Oncol 2000, 39:629-637. tic agents by their selective toxicities toward oxygenated 26. Minn H, Clavo AC, Wahl RL: Influence of hypoxia on tracer and hypoxic tumor cells. Cancer Res 1981, 41:73-81. accumulation in squamous-cell carcinoma: in vitro evalua- 4. Moulder JE, Rockwell S: Hypoxic fractions of solid tumors: tion for PET imaging. Nucl Med Biol 1996, 23:941-946. experimental techniques, methods of analysis, and a survey 27. Reijnders K, English SJ, Krishna MC, Cook JA, Sowers AL, Mitchell JB, of existing data. Int J Radiat Oncol Biol Phys 1984, 10:695-712. Zhang Y: Influence of body temperature on the BOLD effect 5. Overgaard J, Horsman MR: Modification of hypoxia-induced in murine SCC tumors. Magn Reson Med 2004, 51:389-393. radioresistance in tumors by the use of oxygen and sensitiz- 28. Seidel J, Vaquero J, Green MV: Resolution uniformity and sensi- ers. Sem Radiat Oncol 1996, 6:10-121. tivity of the NIH ATLAS small animal PET scanner. Nucl Sci 6. Brown JM: Tumor microenvironment and the response to Symp Conf Rec, IEEE 2001, 3:1555-1558. anticancer therapy. Cancer Biol Ther 2002, 1:453-458. 29. Seidel J, Vaquero J, Pascau J, Desco M, Johnson CA, Green MV: Fea- 7. Khalil AA, Horsman MR, Overgaard J: The importance of deter- tures of the NIH ATLAS small animal PET scanner and its mining necrotic fraction when studying the effect of tumour use with a coaxial small animal volume CT scanner. Biomed volume on tissue oxygenation. Acta Oncol 1995, 34:297-300. Imag Biol 2002, 4:545-548. 8. Krishna MC, English S, Yamada K, Yoo J, Murugesan R, Devasahayam 30. Seidel J, Vaquero JJ, Green MV: Resolution uniformity and sensi- N, Cook JA, Golman K, Ardenkjaer-Larsen JH, Subramanian S, Mitch- tivity of the NIH ATLAS small animal PET scanner: compar- ell JB: Overhauser enhanced magnetic resonance imaging for ison to simulated LSO scanners without depth-of- tumor oximetry: coregistration of tumor anatomy and tis- interaction capability. IEEE Transactions on Nuclear Science 2003, sue oxygen concentration. Proc Natl Acad Sci 2002, 99:2216-2221. 50:1347-1350. 9. Halpern HJ, Yu C, Peric M, Barth E, Grdina DJ, Teicher BA: Oxyme- 31. Johnson CA, Seidel J, Carson RE, Gandler WR, Sofer A, Green MV, try deep in tissues with low-frequency electron paramag- Daube-Witherspoon ME: Evaluation of 3D reconstruction algo- netic resonance. Proc Natl Acad Sci USA 1994, 91:13047-13051. rithms for a small animal PET camera. IEEE Transactions on 10. Velan SS, Spencer RG, Zweier JL, Kuppusamy P: Electron paramag- Nuclear Science 1997, 44:1303-1308. netic resonance oxygen mapping (EPROM): direct visualiza- 32. Krivitski NM, Starostin D, Smith TL: Extracorporeal recording of tion of oxygen concentration in tissue. Magn Reson Med 2000, mouse hemodynamic parameters by ultrasound velocity 43:804-809. dilution. ASAIO J 1999, 45:3-36. 11. Warburg OH: The Metabolism of Tumors London: Constable Publica- 33. Watabe H, Channing MA, Riddell C, Jousse F, Libutti SK, Carrasquillo tions; 1930. JA, Bacharach SL, Carson RE: Noninvasive estimation of the 12. Bentzen L, Keiding S, Horsman MR, Gronroos T, Hansen SB, Over- aorta input function for measurement of tumor blood flow gaard J: Assessment of hypoxia in experimental mice tumours with [O-15]water. IEEE Trans Med Imag 2001, 20:164-174. by [18F]fluoromisonidazole PET and pO2 electrode meas- urements. Influence of tumour volume and carbogen breathing. Acta Oncol 2002, 41:304-312. 13. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P: Asso- ciation between tumor hypoxia and malignant progression Page 9 of 9 (page number not for citation purposes)

Journal

Radiation Oncology – Springer Journals

Published: Feb 28, 2006

There are no references for this article.