| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Medical Oncology (H.W.M.v.L., C.J.A.P.), Nuclear Medicine (L.F.d.G., W.J.G.O.), Surgery (B.W., T.R.), Radiation Oncology (J.L., J.H.A.M.K., A.J.v.d.K.), Radiology (M.R., A.H.), and Medical Technology Assessment (P.F.M.K.), University Medical Center Nijmegen, Geert Grooteplein 8, 6500 HB Nijmegen, the Netherlands. Received August 11, 2004; revision requested October 14; revision received December 1; accepted January 14, 2005. Supported by the Dutch Cancer Society, grant KUN 2000-2307. Address correspondence to H.W.M.v.L. (e-mail: h.vanlaarhoven{at}onco.umcn.nl).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: All patients provided written informed consent, and the study was approved by the institutional review board. A total of 26 patients (12 men and 14 women; mean age, 59 years) who were suspected of having liver metastases of histologically proved colorectal cancer and underwent work-up for liver metastasectomy were included. Patients underwent whole-body FDG PET, and tumor-to-nontumor ratio of FDG uptake in metastases was calculated. Dynamic contrast-enhanced MR imaging was performed, and the rate constant kep (s–1) of gadopentetate dimeglumine uptake in metastases was determined. Pimonidazole was used to determine tumor hypoxia and vascular density of metastases. To assess the relationship between FDG uptake, rate constant kep of gadopentetate dimeglumine uptake, hypoxic fraction, and vascular density, the Pearson correlation coefficient was calculated.
RESULTS: Negative correlation between tumor-to-nontumor ratio of FDG uptake and rate constant kep was observed (r = –0.421, P = .082). No correlation between tumor hypoxia and tumor-to-nontumor ratio of FDG uptake or rate constant kep was found. A positive correlation was observed between vascular density and rate constant kep (r = 0.458, P = .034) but not between tumor-to-nontumor ratio of FDG uptake.
CONCLUSION: Negative correlation between tumor-to-nontumor ratio of FDG uptake and rate constant kep suggests that lower values of gadopentetate dimeglumine uptake imply an acutely reduced supply of oxygen, which necessitates a higher uptake of glucose to maintain tumor energy levels. The positive correlation of vascular density with rate constant kep, but not with tumor-to-nontumor ratio of FDG uptake, emphasizes the potential of dynamic contrast-enhanced MR imaging to enable measurement of tumor vascularity in vivo and its additional value compared with ex vivo methods.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Although FDG uptake generally is increased in malignant tumors, the underlying mechanism for increased glucose uptake, and thus for increased FDG PET signal, is still a matter of debate. For glucose to be taken up and used by a cancer cell, an adequate vascular supply—and thus angiogenesis or vascular cooption—is necessary, as is the presence of several membrane-bound glucose transport proteins, which facilitate transport of glucose over the cell membrane. The intracellular hexokinase isoforms are necessary for subsequent phosphorylation into glucose-6-phosphate, which may then be further converted via the glycolytic pathway. All of these factors (ie, vascularity, transmembrane transport, phosphorylation, and glycolysis) are known to be upregulated in cancer cells (7); this may be due to overexpression of the hypoxia-inducible factor 1
protein. Hypoxia-inducible factor 1 can be overexpressed in tumors in response to tumor hypoxia resulting from an inefficient tumor vascular network and constitutively as a result of gene mutations (8). In fact, (constitutive) overexpression of hypoxia-inducible factor 1 induces vascular endothelial growth factor expression, which may lead to an inefficient tumor vascular network and result in insufficient tumor oxygenation. Thus, tumor hypoxia seems to play a pivotal role in the metabolic status of tumors.
To improve the understanding of PET findings, several studies have focused on the link between FDG uptake and biomarkers such as glucose transport proteins, hexokinase, and vascular density. Some authors have reported positive correlations between these biomarkers and FDG uptake (9,10), but other authors have reported negative correlations (11–15). Divergent results may be explained by differences in tumor biology, immunohistochemical staining methods, and PET procedures. However, one usually does not account for the fact that FDG uptake is measured in vivo, whereas ex vivo data are obtained with immunohistochemical analysis. This may be particularly important for the (lack of a) relationship between FDG uptake and vascular density, since it is not so much the presence of tumor vasculature but the presence of functional tumor vasculature that may play a role in the determination of FDG uptake.
Functionality of tumor vasculature can be monitored in vivo by using dynamic contrast material–enhanced magnetic resonance (MR) imaging with gadopentetate dimeglumine. From physiologic models for the analysis of dynamic contrast-enhanced MR imaging, data parameters for vascularity (eg, blood flow and permeability of blood vessels) can be determined (16,17). Dynamic contrast-enhanced MR imaging currently is used for tumor identification in the clinic. For example, in patients with breast cancer, dynamic contrast-enhanced MR imaging has been proved to be an accurate method in the differentiation of benign and malignant lesions (18). Dynamic contrast-enhanced MR imaging parameters may provide a useful noninvasive measure in both the prediction of treatment outcome and the follow-up of therapy (19); furthermore, they have been shown to have predictive value in the response to treatment of several primary tumors (20–25).
The purpose of our study was to examine the in vivo relationship between FDG uptake, as measured with PET, and functional tumor vasculature, as measured with dynamic contrast-enhanced MR imaging, in patients with liver metastases of colorectal cancer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
FDG PET Scanning
Whole-body FDG PET scanning was performed as part of the work-up for liver metastasectomy when clinically indicated according to the treating surgeon. Imaging was performed with a dedicated PET scanner (Ecat Exact; Siemens/CTI, Knoxville, Tenn). Prior to FDG injection, patients fasted for at least 6 hours. Intake of sugar-free liquids was permitted. Immediately before the procedure, the patients were hydrated with 500 mL of water. One hour after intravenous injection of 200–220 MBq FDG (Mallinckrodt Medical, Petten, the Netherlands) and 12 mg furosemide, emission and transmission images of the area between the proximal femora and the base of the skull were acquired (10 minutes per bed position). The images were corrected for attenuation and reconstructed by using the ordered-subsets expectation maximization algorithm.
FDG PET Image Analysis
Tumor metabolism was evaluated semiquantitatively with calculation of the tumor-to-nontumor ratio of FDG uptake by using a semiautomated method, with normal liver tissue adjacent to the lesions serving as the reference standard. A volume of interest in liver metastasis was defined by using a 50% threshold of maximum intensity. Central photopenic areas in the metastases, which may be regarded as areas of gross tumor necrosis, were excluded from the volume of interest. The images were analyzed by two independent observers (W.J.G.O. and L.F.d.G., with 7 and 4 years of PET experience, respectively) who were blinded to findings of quantitative MR imaging and histologic analysis. Disagreements between the two observers were resolved by consensus. The normal liver volume of interest was placed adjacent to the measured lesion. The boundaries of the volume of interest were just within the apparent hypermetabolic zone of the tumor. Volumes of interest of identical configuration were placed on normal liver tissue to serve as the reference standard for normalization.
Dynamic Contrast-enhanced MR Imaging
Dynamic contrast-enhanced MR imaging was performed in all patients who underwent work-up for liver metastasectomy as part of a wider research program to validate the use of dynamic contrast-enhanced MR imaging in patients with liver metastases. Three authors (H.W.M.v.L., M.R., and A.H., all with at least 4 years of experience with liver MR) were involved in dynamic contrast-enhanced MR imaging and data collection. There was no randomization as to which test (MR imaging or PET) was performed first; the clinical order was followed. In practice, patients underwent PET before or on the same day as MR imaging.
Examinations were performed with a 1.5-T MR system (Vision; Siemens) and a body phased-array coil. After conventional T1- and T2-weighted imaging in the transverse, coronal, and saggital directions, a 15-mL dose of 0.5 mol/L gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) was administered intravenously with an MR injection system (Spectris; Medrad, Maastricht, the Netherlands) and an injection rate of 2.5 mL/sec. Gadopentetate dimeglumine uptake in the tumor and bolus passage in vessels in the spleen were monitored by using a T1-weighted fast low-angle shot sequence with a time resolution of 2 seconds. Sequence parameters were as follows: repetition time msec/echo time msec, 50.0/4.4; flip angle, 90°; section thickness, 7 mm; four sections acquired; matrix, 160 x 256; field of view, 263 x 350 mm. Dynamic contrast-enhanced MR imaging data were acquired for 90 seconds. If the four sections did not fully cover the tumor in the head-foot direction, sections were positioned in such a way that the largest diameter of the tumor (measured left to right on the coronal image) was covered.
Just before gadopentetate dimeglumine injection, proton-density weighted images were obtained with the same sequence parameters as were used for dynamic contrast-enhanced MR imaging, with the exception of an 8° flip angle and a 200-msec repetition time. Data from these images were combined with the dynamic contrast-enhanced MR imaging data to calculate the concentration of gadopentetate dimeglumine in arbitrary units by using the method described by Hittmair et al (26).
For analysis of dynamic contrast-enhanced MR imaging data, to which readers were blinded for analysis of PET data, we used the previously described method (27). In brief, we obtained a vascular normalization function from pixels in the spleen by using an algorithm based on the concentration of gadopentetate dimeglumine (which is high in blood vessels) and time to bolus passage (which is short in arteries). A physiologic pharmacokinetic model (17) was used to analyze the gadopentetate dimeglumine concentration versus time curves of the pixels in all MR sections containing tumor tissue. Values of the rate constant kep (s–1) of gadopentetate dimeglumine uptake were calculated with the following formula: Ct(t) = Ktrans · e–kep·t * Cp(t), where Ct(t) = tissue concentration of gadopentetate dimeglumine as a function of time, Ktrans = volume transfer constant (s–1), kep = rate constant (s–1) between extravascular extracellular space and blood plasma, and Cp(t) = concentration of contrast agent in plasma of a capillary as a function of time; * denotes a convolution operation (28). In the model of Larsson et al (17), the gadopentetate dimeglumine uptake rate constant kep is directly related to tumor blood flow, the product of the permeability of perfused capillaries, and the total surface area of perfused capillaries, according to the following equation: kep = (1 – exp[–PS/TBF]) · TBF/Ve, in which Ve = volume of contrast extravascular extracellular space per unit volume of tissue, P = permeability of capillaries (measured in centimeters per second), S = total surface area of vessels (measured in square centimeters), PS = permeability surface area product (measured in milliliters per second), and TBF = tumor blood flow (measured in milliliters per second).
The spatial distribution of the kep values was represented in a map. On a T1-weighted MR image obtained directly before gadopentetate dimeglumine injection, a region of interest that comprised the metastases was drawn. This region of interest was applied to the map of the rate constant kep of gadopentetate dimeglumine uptake to enable selection of the single kep values for all tumor pixels. The mean of the rate constant kep of these pixels was calculated after log transformation and was averaged over all sections containing tumor tissue. Log transformation excludes all values of kep equal to zero. These tumor pixels may be regarded as necrotic tumor parts. Back transformation of this average log-transformed value resulted in an average kep value for the whole tumor (27).
In 12 patients for whom sufficient time remained before surgery was planned, the measurement protocol was repeated with an interval of 1–4 days as part of a reproducibility study. The dynamic contrast-enhanced MR imaging results of 10 of these 12 patients have been published elsewhere (27). For these patients, the average kep value for the whole tumor was calculated from the mean of the two measurements, and this mean value was used for further calculations.
Tumor Hypoxia and Vascular Density
Pimonidazole (1-[(2-hydroxy-3-piperidinyl)propyl]-2-nitroimidazole hydrochloride; Natural Pharmacia International, Belmont, Mass) was selected as a hypoxia marker (29–31) and injected intravenously over 20 minutes at a dose of 500 mg per kilogram of body weight at least 12 hours before the start of surgery. Pimonidazole is a bioreductive chemical probe with an immunorecognizable side chain. Complete inhibition of bioreductive activation occurs at a partial pressure of oxygen of less than 10 mm Hg (31).
Immediately after surgical resection, the metastasis for which dynamic contrast-enhanced MR imaging data were available was identified in the gross pathologic specimen by using the T1- and T2-weighted MR images obtained in the transverse, coronal, and saggital directions. On the T1-weighted MR images, the distance in the craniocaudal direction was measured from the edge of the metastasis to the center of the region where the sections for dynamic contrast-enhanced MR imaging were obtained. In the gross pathologic specimen, the same distance was measured; at this level, a pathologist cut a 3-mm transverse slice of the liver metastasis. In case of relatively small metastases (approximately 2 cm), the whole slice was taken for further analysis. In case of larger metastases maximally, five smaller biopsy samples (approximately 5 x 5 mm) were taken from the larger slice (four biopsy samples from the rim and one from the center). The biopsy samples from the rim were evenly distributed (ie, one from the dorsal part, one from the ventral part, one from the lateral part, and one from the medial part of the tumor). The material was snap-frozen in isopentane (BDH, Dagenham, England), precooled in liquid nitrogen, and stored at –80°C until further use. The differentiation grade and largest size of the metastases were recorded from the clinical report of gross pathologic analysis.
Frozen tumor slices of 5-µm thickness were cut for immunohistochemical staining and analysis of hypoxia and vascular density. After thawing, the slices were fixed in cold (4°C) acetone for 10 minutes and rehydrated in phosphate-buffered saline for 30 minutes. Between the consecutive steps of the staining procedure, the slices were rinsed three times for 2 minutes in phosphate-buffered saline. To stain the hypoxic marker and the vasculature, slices were incubated with a mouse antibody (Pathologie Anatomie Leiden-Endotheel, Department of Pathology, Leiden University, Leiden, the Netherlands) that was diluted to a 1:15 ratio and with a rabbit antipimonidazole (Department of Radiation Oncology, North Carolina Clinical Cancer Center, Chapel Hill, NC) (30,32) that was diluted to a 1:200 ratio in polyclonal liquid dilutant (Euro DPC, Breda, the Netherlands) during 30 minutes at 37°C. Slices were then incubated with goat antimouse TexasRED (Jackson Immuno Research Laboratories, West Grove, Penn) and donkey antirabbit Alexa488 (Molecular Probes, Leiden, the Netherlands), both diluted to a 1:200 ratio in polyclonal liquid dilutant during 30 minutes at 37°C. Finally, slices were mounted in Fluorostab (Organon, Boxtel, the Netherlands).
Quantitative data for hypoxia and tumor vasculature were acquired with a semiautomatic method based on a computerized digital image analysis system, as described previously (33,34). In brief, a high-resolution intensified solid-state camera on a fluorescence microscope (Axioskop; Zeiss, Weesp, the Netherlands) with a computer-controlled motorized stepping stage was used to examine each tumor sample. Whole tumor samples were examined at x100 magnification with different filters for the detection of the fluorescent signals. Each examination consisted of 36–144 fields of 1.2 mm2, depending on the size of the tumor slice. From the individual microscopic fields, one composite image was reconstructed after each examination. As a final step, a contour line was drawn to delineate the viable tumor area by using consecutive hematoxylin-eosin stained tumor slices to distinguish tumor tissue from nontumor tissue. The hypoxic fraction of the tumor slice was computed as the tissue surface area stained by the hypoxic marker relative to the viable tumor surface area, and vascular density was calculated as the total number of vessels per square millimeter of viable tumor area. The average hypoxic fraction and vascular density for each liver metastasis were calculated, averaging the scores of all tumor sections per liver metastasis.
Statistical Analysis
The mean and range of tumor-to-nontumor ratio, rate constant kep, hypoxic fraction, and vascular density for the whole population were calculated. To assess the relationship between FDG uptake, rate constant kep of gadopentetate dimeglumine uptake, hypoxic fraction, and vascular density, Pearson correlation coefficients were calculated with computer software (SPSS, version 12.0.1; SPSS, Chicago, Ill). To calculate the 95% regression coefficient and prediction intervals for the relationship between the FDG tumor-to-nontumor ratio and rate constant kep, SigmaPlot version 9 software (Systat Software, Richmond, Calif) was used. P values of less than .05 were indicative of a significant difference, whereas P values of less than .1 were indicative of a trend.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
Comparison of Tumor-to-Nontumor Ratio and Rate Constant kep
From the FDG PET images (Fig 1a), the tumor-to-nontumor ratio was calculated for all patients for whom FDG PET data were available (ie, 19 of 26 patients). The mean tumor-to-nontumor ratio of these patients was 2.021 ± .187. On the T1-weighted MR image recorded just before gadopentetate dimeglumine administration, the metastases could easily be detected in all patients for whom dynamic contrast-enhanced MR imaging data were available (ie, 25 of 26 patients) (Fig 1b). The mean rate constant kep of these patients was 0.031 (standard error of the mean < .001). Comparison of the tumor-to-nontumor ratio and rate constant kep showed a trend for higher tumor-to-nontumor ratios at lower rate constant kep values (P = .082; Fig 2). Excluding one observation with a relatively high tumor-to-nontumor ratio (upper open circle in Fig 2) from the analysis resulted in a significant correlation between the tumor-to-nontumor ratio and rate constant kep (P = .049). The omitted patient did not differ from the whole population with respect to age, comorbidity, medication, differentiation grade of the metastasis, or tumor diameter, as measured after resection.
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Our results showed a negative correlation between tumor-to-nontumor ratios of FDG uptake and gadopentetate dimeglumine uptake rate constants kep in liver metastases. When tumor blood flow is much greater than permeability surface area product, lower values of rate constant kep may indicate a lower permeability surface area product (see formula for calculating rate constant kep in Materials and Methods). When tumor blood flow is much less than permeability surface area product, lower values of rate constant kep may indicate lower tumor blood flow. A lower tumor blood flow, lower permeability, or smaller surface area of tumor blood vessels may all result in a reduced supply of nutrients, such as FDG and oxygen, to the tumor. A reduced supply of nutrients could lead to lower FDG tumor-to-nontumor ratios in the tumor, as was shown in an experimental study in which substrate availability modulated glucose metabolism (35) and would result in a positive correlation between tumor-to-nontumor ratios and rate constant kep. The poorer vascular function, which was reflected by lower values of rate constant kep, however, could lead not only to a reduced supply of nutrients but also to a reduced supply of oxygen. A reduced supply of oxygen would necessitate a higher uptake of glucose to maintain tumor energy levels. This would lead to higher FDG tumor-to-nontumor ratios in the tumor, thus resulting in a negative correlation between rate constant kep and FDG tumor-to-nontumor ratios. Thus, the observed negative correlation between rate constant kep and FDG tumor-to-nontumor ratios suggests that differences in tumor oxygenation rather than differences in FDG delivery are a driving force for FDG uptake in colorectal liver metastases.
Whether FDG delivery or tumor oxygenation are the driving force for FDG uptake may be tumor and even stage dependent. For example, a positive correlation between FDG uptake and the permeability surface area product, as determined with dynamic contrast-enhanced MR imaging, was found in patients with lung cancer (36) and that between the metabolic rate of FDG and blood flow determined with oxygen 15 water measurements was found in patients with advanced breast cancer (37). Brix et al (38) did not observe a correlation between FDG PET scanning and dynamic contrast-enhanced MR imaging parameters in a group of patients with suspicious breast lesions. Clearly, this group consisted of patients with different stages of breast cancer, which might explain the lack of correlation between FDG PET scanning and dynamic contrast-enhanced MR imaging parameters.
Relation between Hypoxic Fraction, FDG Tumor-to-Nontumor Ratio, and Rate Constant kep
The hypothesis that tumor hypoxia is a driving force for FDG uptake in colorectal liver metastases seems to contradict the observed lack of correlation between FDG uptake and hypoxic fraction in liver metastases, as measured with pimonidazole binding. In this respect, the difference between chronic or diffusion-limited hypoxia, according to the classic model of Thomlinson and Gray (39), and acute or transient hypoxia (40) because of local and temporary fluctuations of tumor blood perfusion may be relevant. Experimental tumor cells in an acutely hypoxic environment may increase their FDG uptake more than twofold (41,42) to survive the temporary decrease in oxygen supply. Decreased cell proliferation has been shown in chronically hypoxic regions (43). This may be interpreted as an energy-saving method to adapt to a reduced supply of oxygen and nutrients. Alternatively, it may be argued that an important part of the chronically hypoxic cell population is becoming necrotic or apoptotic and is, therefore, less active metabolically. In these chronically hypoxic regions, FDG uptake will actually decrease. Thus, hypoxic tumors can be highly metabolic or may have modest glucose metabolism, depending on whether acute or chronic hypoxia plays a major role (44). The discordance of FDG uptake and tumor hypoxia can be specific to tumor type (44), and it may even be heterogeneous within one tumor (45). Both chronically and acutely hypoxic cell regions contribute to the hypoxic fraction measured with pimonidazole binding. This may explain the lack of correlation between FDG uptake and hypoxia in liver metastases in our study.
Relation between Vascular Density, FDG Tumor-to-Nontumor Ratio, and Rate Constant kep
We observed a positive correlation between vascular density and the gadopentetate dimeglumine uptake rate constant kep but not between vascular density and FDG uptake. A positive correlation between microvascular density and dynamic contrast-enhanced MR imaging vascular parameters has been reported by Hawighorst et al (25) for cervical carcinoma, but Su et al (46) did not observe such a correlation in patients with breast cancer. As explained previously, rate constant kep is defined both by the product of permeability and surface area of perfused capillaries and by the tumor blood flow. From this definition, the relationship between the gadopentetate dimeglumine uptake rate constant kep and the vascular density is obvious. It should be noted, however, that in vascular density, both perfused and nonperfused vessels are included. Since mainly the acute closure of the former seems to determine uptake of FDG in colorectal liver metastases, this may explain the lack of correlation between vascular density and FDG uptake, and it underscores the importance of functional in vivo data on the tumor vascular system.
Limitations of the Study
The mean interval between FDG PET and dynamic contrast-enhanced MR imaging and FDG PET and surgery was relatively long in our study. It may be suggested that the lack of correlation between FDG PET and immunohistochemical results was caused by this long interval. However, since it seems unlikely that the correlation between FDG PET and dynamic contrast-enhanced MR imaging data are caused by the long interval between the two examinations, the lack of correlation between FDG PET and immunohistochemical results should be explained by biologic factors, rather than the long interval. Nevertheless, its influence cannot be excluded. Two patients received anticancer therapy between FDG PET scanning and dynamic contrast-enhanced MR imaging. Although this may have influenced the results, it pertains to a small number of patients.
The mean value of rate constant kep indicates the value of this parameter over the region of interest, but it does not reflect the heterogeneity of a tumor. Currently, statistical analyses, such as functional principal component analysis, are being investigated to further characterize spatial heterogeneity (47). Although tumor heterogeneity was not taken into account in our study, all data are corrected for (gross) necrosis.
It should be noted that an exact one-to-one correlation between FDG PET or dynamic contrast-enhanced MR imaging and immunhistochemistry data was not fully possible, since tissue sections had a size of approximately 5 µm x 5 mm x 5 mm, whereas PET and MR imaging data were obtained from the whole tumor or a large part of the tumor. Also, in two patients, the metastases were not fully removed, and biopsy was performed during surgery. In our study, we assumed that the acquired material was a sufficient representation for the whole tumor. Since the sample size of this study was small, it should be expanded in a future study.
For colorectal liver metastases, a negative relationship was found between the rate constant kep of gadopentetate dimeglumine uptake, as determined with dynamic contrast-enhanced MR imaging, and FDG tumor-to-nontumor ratios, as determined with PET scanning. This suggests that in liver metastases, differences in acute tumor hypoxia—which are caused by differences in functional tumor vasculature—constitute a driving force for FDG uptake. The observed correlation between vascular density and the rate constant kep of gadopentetate dimeglumine uptake, but not with tumor-to-nontumor ratios, emphasizes the potential of dynamic contrast-enhanced MR imaging to enable measurement of tumor vascularity in vivo and its additional value compared with ex vivo methods.
|
FOOTNOTES |
|---|
Abbreviations: FDG = fluorine 18 fluorodeoxyglucose
2 Current address: Department of Psychonomics, University Utrecht, the Netherlands 
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, H.W.M.v.L.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, H.W.M.v.L., L.F.d.G., B.W., M.R., J.H.A.M.K., A.J.v.d.K., W.J.G.O., A.H.; clinical studies, H.W.M.v.L., L.F.d.G., B.W., J.L., J.H.A.M.K., T.R.; statistical analysis, H.W.M.v.L., P.F.M.K.; and manuscript editing, H.W.M.v.L., L.F.d.G., B.W., M.R., J.H.A.M.K., P.F.M.K., C.J.A.P., W.J.G.O., A.H.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Bisdas, K. Spicer, and Z. Rumboldt Whole-Tumor Perfusion CT Parameters and Glucose Metabolism Measurements in Head and Neck Squamous Cell Carcinomas: A Pilot Study Using Combined Positron-Emission Tomography/CT Imaging AJNR Am. J. Neuroradiol., August 1, 2008; 29(7): 1376 - 1381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Miles, C. Keith, H. W. M. van Laarhoven, L.-F. de Geus-Oei, B. Wiering, J. Lok, M. Rijpkema, J. H. A. M. Kaanders, P. F. M. Krabbe, T. Ruers, et al. Gadopentetate Dimeglumine and FDG Uptake in Liver Metastases Radiology, October 1, 2006; 241(1): 319 - 320. [Full Text] [PDF]
|
|
|
|
|
|
A. H. Khandani and R. C. Semelka Can the Relation between Gadopentetate Dimeglumine and FDG Uptake in Colorectal Liver Metastases Be Used Clinically? Radiology, October 1, 2005; 237(1): 1 - 2. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
