BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
British Journal of Radiology (2004) 77, 296-307
© 2004 British Institute of Radiology
doi: 10.1259/bjr/95415645
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dzik-Jurasz, A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dzik-Jurasz, A

Review article

The development and application of functional nuclear magnetic resonance to in vivo therapeutic anticancer research

2002 Sir Godfrey Hounsfield lecture delivered at the President's Day, Manchester

A Dzik-Jurasz, PhD, FRCS, FRCR

The Cancer Research UK Clinical Magnetic Resonance Research Group, Institute of Cancer Research and the Royal Marsden NHS Trust, Sutton, Surrey SM2 5PT, UK

Correspondence: Dr A Dzik-Jurasz, GlaxoSmithKline Pharmaceuticals, 891-995 Greenford Road, Bldg 5 Flr 1 Rm 13, London UB6 0HE, UK


   Abstract
Top
 Abstract
Introduction
 Multifunctional studies in...
 Developing non-invasive...
 References
 
A little over 30 years ago, Sir Godfrey Hounsfield and his colleagues revolutionized medical imaging by developing CT scanning. In recent years a combination of improved technology and a deeper understanding of tumour biology have led to the development of imaging based strategies aimed at interrogating tissue structure and function. The prospects of this new technology include the prediction of tumour response and the non-invasive study of conventionally inaccessible yet important pharmacological compartments. This article explores how functional nuclear MRI and spectroscopy have been used in predicting response to anticancer therapy in rectal cancers and to assess the biliary excretion of chemotherapeutics.


   Introduction
Top
 Abstract
 Introduction
Multifunctional studies in...
 Developing non-invasive...
 References
 
In his 1979 Nobel speech (available at http://www.nobel.se/medicine/laureates/1979/), Sir Godfrey Hounsfield (Figure 1Go) described the principles of computerized axial tomography including the reconstruction strategies required to display the images. It might however surprise readers to find the description of another up and coming imaging modality at that time, namely that of nuclear magnetic resonance (NMR) imaging (Figure 1Go). It was Sir Godfrey's insight even then that led him to articulate the potential of probing not only the morphological but also the biophysical character of tissues.



View larger version (63K):
[in this window]
[in a new window]
  		Figure 1. Sir Godfrey Hounsfield (left) and the title of his 1979 Nobel lecture (top right) together with a diagram illustrating the variation in precessional frequency in nuclei exposed to a magnetic gradient orthogonal to the main field (bottom right) from the same lecture. © The Nobel Foundation.

 
The limitation of morphological imaging [1] in reflecting underlying tumour behaviour is perhaps best illustrated in studies of the tumour microenvironment. Tumour biology [2] is typically heterogeneous both spatially and temporally [3]. The heterogeneity is further compounded through interplay of tumour and host factors with additional modulation via any number of administered therapies. It is also the case that organism survival is more dependent on cell kill and repopulation kinetics than the fraction of tumour cells killed [4]. Thus, morphological features often poorly reflect underlying tumour biology.

I would like to describe two sets of work carried out by my colleagues and myself over the last 5 years, which have exploited the capability of NMR to probe the biophysical environment of fluids and tissues in vivo. The NMR strategies we used comprised an assessment of tumour vascularity via permeability and perfusion studies, an assessment of tissue architecture by quantitative diffusion weighted imaging (DWI), and studies of tumour biochemistry via 1H-MRS (magnetic resonance spectroscopy). The application of MRS to monitoring drug metabolism is pursued in relation to our study of biliary excretion of the catabolite of an anticancer agent via resonance of its 19F nucleus.

A diverse set of physical and metabolic mechanisms amalgamate to express the tumour phenotype and our aim was to identify novel predictors of response using biophysical parameters measurable via NMR. Our focus on rectal cancer is clinically relevant since it remains the second most common malignancy of adults in the UK. It was also considered that such a study would complement other studies by a member of our department which have demonstrated a close correlation between NMR imaging of rectal tumours and histology [5]. In addition, with the advent of molecular (cell receptor and gene therapies) medicine there is a real prospect of individualized patient treatment.


   Multifunctional studies in locally advanced rectal cancer
Top
 Abstract
 Introduction
 Multifunctional studies in...
Developing non-invasive...
 References
 
Vascular characteristics
We studied patients with locally advanced rectal adenocarcinoma by multifunctional NMR imaging. All patients received a 12-week chemotherapy module followed by 8 weeks of combined chemotherapy and radiotherapy. The functional NMR studies were performed prior to, at the end of chemotherapy and following the end of chemoradiotherapy. Functional studies are aimed at extracting physiological information on the basis of inherent or modulated NMR signals. In studies reflecting contrast delivery and leakage, a pixel-by-pixel estimate of the T1 (calculated from the ratio of a proton density to T1 weighted image) is converted to a gadolinium concentration–time curve. The gadolinium concentration–time curve is then fitted to a choice of mathematical models from which representative physiological parameters such as Ktrans (volume transfer constant between blood plasma and extravascular extracellular space), Ve (volume of extravascular extracellular space per unit volume of tissue), and kep (rate constant between the extravascular extracellular space and blood plasma) are derived [6]. Perhaps the best-known mathematical model is that due to Tofts and Kermode and this was the model applied to our rectal data. Although the acquisition of a gadolinium concentration–time curve uses the well-recognised NMR phenomenon of T1 enhancement the basis for the acquisition of blood volume data is perhaps not so well recognised. In its first pass through a vessel or region of interest a paramagnetic agent causes an alteration in the local magnetic gradients that manifest as susceptibility (Figure 2Go). The resulting contrast in an appropriately weighted T2* sequence therefore manifests as signal loss. The numerical value of the area defined by the signal loss is theoretically related to blood volume. Since the value of the area defined by the signal loss quantification is not absolute, the measure is termed "relative blood volume". In our study we used a gradient echo sequence that provided single slice T1 and T2* data [7]. We found that there was a linear correlation between the Ktrans, a measure of capillary leakiness, prior to treatment and response to chemotherapy (Figure 3Go) as measured according to World Health Organization (WHO) criteria [8]. Thus a tumour with a higher mean value of capillary leakiness (Figure 4a, bGo) was more likely to respond than one with a lower mean value (Figure 4c, dGo). Importantly, this finding was irrespective of tumour size. Interestingly, a further linear correlation was found between the percentage of visible signal loss in the tumour slice (relative blood volume) and response to chemotherapy again irrespective of tumour size. Therefore a tumour with a greater relative blood volume (Figure 5a, bGo) would be more likely to respond to chemotherapy than one with a smaller or unrecordable relative blood volume (Figure 5c, dGo).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 2. Behaviour of NMR signal intensity in response to a bolus intravenous dose of paramagnetic contrast agent. The signal intensity was derived from the tumour that had previously been outlined. The T1 weighted sequence records a rise in signal intensity due to shortening of the spin-lattice relaxation time of water protons. With a T2* weighted sequence there is a transient loss of signal intensity due to the sequence detecting a first pass effect of the contrast agent. The rapid change in the local magnetic fields induced by the first pass of contrast agent induces a strong susceptibility effect. The area under the T2* curve is theoretically related to the blood volume in the region of interest.

 


View larger version (14K):
[in this window]
[in a new window]
  		Figure 3. Association of the measured capillary permeability in the locally advanced rectal cancers in our study with response following chemotherapy. The association indicates that tumours exhibiting a greater capillary permeability (as measured by Ktrans) are more likely to respond to treatment than those that are not. If this association does reflect underlying biology then it may be related to drug delivery. Alternatively it may simply be reflecting a particularly susceptible phenotype. Ktrans is expressed as the logarithm since the transform resulted in a normal distribution of the data.

 


View larger version (124K):
[in this window]
[in a new window]
  		Figure 4. Example of two locally advanced rectal tumours with (a), (b) high and (c), (d) low permeability. The top images are the anatomical axial T2 images whilst the lower two demonstrate a "permeability" (capillary leakiness) map overlaying the two tumours. The colour scale is shown on the left of images (b) and (d) and permits an appreciation of the extent of capillary leakiness within individual voxels based on modelling kinetic contrast agent data (in this case Gd-DTPA, gadolinium(III) diethylenetriamine pentaacetic acid). Our results indicate that rectal tumours with a higher permeability (b) are more likely to respond to chemotherapy than those with low permeability (d). Our findings are independent of original tumour size.

 


View larger version (124K):
[in this window]
[in a new window]
  		Figure 5. Tumours with a greater relative blood volume were more likely to respond to chemotherapy than those with a lower or no recordable blood volume. The top images (a and c) are the T2 weighted axial anatomical images of the tumours and the lower two images are the relative blood volume or "susceptibility" maps. The "susceptibility" maps arise from a first pass effect of contrast agent through the tumour inducing a susceptibility effect. Although the physical effect is manifest as a signal loss, the scale has been inverted to highlight the areas of signal loss. The tumour illustrated on the left (a and b) demonstrates a strong "susceptibility" effect whilst there was no recordable signal loss in the tumour on the right (c and d). The tumour on the right is less likely to respond to chemotherapy than the one on the right.

 
Diffusion characteristics
Diffusion describes the migration of molecules down a concentration gradient. The process is random and results, for example, in the equal distribution of a solute in solution, i.e. an ink drop in jar of water. With the appropriate choice of local magnetic gradients it is possible to weight an NMR image such that the recorded signal intensity is a function of the translational (diffusion) motion of the nucleus of interest. The self-diffusion coefficient of compounds has been measured via NMR for over four decades. The more complex situation encountered in vivo where more than one environment is often present results in the measure of a weighted sum of diffusion coefficients termed the apparent diffusion coefficient (ADC). The added issues of susceptibility and physiological motion make the matter technically more complicated in vivo. Although the technique is most commonly used for water protons because of the NMR sensitivity of the nucleus, the diffusion of any other NMR visible nucleus is possible given sufficient sensitivity [9].

In the rectal study we used a technique developed for fast imaging tailored to quantitative diffusion measurements. The sequence was based on the Burst strategy [10] in which a train of small flip angle pulses are applied with the signal being sampled during a constant read gradient in between which time a non-selective 180° pulse is applied. The result is a train of 16 images with b-values (this is essentially a factor that represents how sensitive a sequence will be to diffusion effects. It is common to all diffusion weighted sequences) varying between 0 and 396 s mm–2. The slope of a semilogarithmic plot of signal intensity against b-value gave the mean ADC of the tumour outlined on a low b-value image. A plot of the mean tumour ADC prior to treatment against response to chemotherapy and chemoradiotherapy (Figure 6Go) demonstrated a strong linear correlation [11]. We found that those tumours with a high ADC were far less likely to respond to chemotherapy and chemoradiotherapy than those with an initially low ADC. It is known from animal work that the ADC in experimental tumours is often associated with the extent of necrosis [12, 13] and on this basis we have hypothesised that the mean ADC measured in the rectal tumours is a surrogate marker of necrosis. In other words, necrotic tumours are less likely to respond to treatment than those that are not. In itself this is not a novel interpretation since the negative influence of necrosis on treatment has long been recognised [14]. I would suggest that the novelty of the finding is in the ability to predict, prior to treatment, which patients are likely to respond and which will not. This might in the future allow patients to be stratified individually prior to treatment. It is worth adding that several works have reported that an increase in ADC indicates early response in animal tumour models treated by conventional [12, 15–18] or gene therapy [19, 20] and similar behaviour has been reported in a limited number of human brain tumours [21, 22]. However these studies report on changes which have occurred after treatment has started.



View larger version (85K):
[in this window]
[in a new window]
  		Figure 6. Results of studying locally advanced rectal cancer using quantitative diffusion weighted imaging. A typical low-b value image (top left) of a large tumour and the resulting diffusion image (top right). The two graphs demonstrate that those tumours with a low apparent diffusion coefficient (ADC) (and by implication of higher cellularity) prior to treatment are more likely to respond to chemotherapy (bottom left) and chemoradiation (bottom right) than those that are not. Based on evidence from animal tumours we have argued that the ADC measurements in this situation are a surrogate marker for necrosis. Graphs reprinted with permission from Elsevier [11].

 
Correlation with disease biomarkers
Since functional NMR studies purport to represent mechanisms occurring at the cellular/molecular level we considered it was important where possible to determine whether there were associations between disease biomarkers and our measured values of capillary leakiness. There are a considerable number of vasoactive peptides involved in the angiogenic [23] process many of which are upregulated in cancer. We therefore hypothesised that if our NMR derived values of capillary leakiness were reflective of the underlying angiogenic process then we should at least expect an association between some of the measured variables. Essentially, we detected no evidence of a correlation between the mean Ktrans value of tumours pre-treatment and carcinoembryonic antigen (CEA) or serum levels of basic fibroblastic growth factor (bFGF). However, a significant correlation (Figure 7Go) was detected between the measure of capillary leakiness (Ktrans) and serum vascular endothelial growth factor (VEGF) levels [24]. CEA is a tumour cell antigen involved in intercellular adhesion and expressed in approximately 85% of patients with colorectal cancer [25]. CEA does not significantly influence the angiogenic process and therefore it is unsurprising that no correlation was found. The vascular peptide bFGF however plays a central role in angiogenesis via effects on endothelial cells, pericytes and matrix degradation and production. It does not, however, significantly affect vessel permeability whereas VEGF is highly potent in this respect. We have argued that if our NMR measure of capillary leakiness were reflecting the underlying angiogenic process then an association should be expected with VEGF but not bFGF or CEA. Although the considerable clustering of data points at the lower end of the graph in Figure 7Go is acknowledged, a linear relationship within the data exists nevertheless. This finding argues that some biological relationship might exist between our NMR measures of capillary leakiness and serum VEGF levels (and by implication VEGF production). We recognise, however, that other confounding factors may be contributing to our findings. It is interesting to note that other in vivo NMR studies have demonstrated associations between NMR measures of vascularity and markers of angiogenesis [26] albeit with varying methodology.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 7. Association between the measure of capillary permeability (Ktrans) and the serum vascular endothelial growth factor (VEGF) platelet ratio. The association supports, though far from proves, that MR based measures of vascularity, in this case capillary permeability are related to the underlying angiogenic process. Platelets normally carry VEGF and therefore we were concerned to account for the patient's platelet count, hence the expression of the ratio.

 
Findings at 1H-magnetic resonance spectroscopy
Multinuclear MRS has proved to be a particularly powerful research tool [27] in the study of endogenous biochemistry [28] and in vivo anticancer drug metabolism [29]. The direct clinical impact of MRS on oncological practice has however been limited. This is probably due to the limited training of clinical radiologists in the field and the additional costs of spectroscopic equipment especially when nuclei other than 1H are concerned. We were able to record 1H-spectra from 6 out of 21 (29%) of the patients with locally advanced rectal cancer [30]. Only six tumours had a morphology that accommodated a 2 cm3 voxel. Although all the tumours had volumes in considerable excess of 2 cm3 the presence of luminal air (which significantly affects spectral linewidth and hence spectral resolution) severely restricted voxel placement. Those spectra available for analysis (Figure 8Go) demonstrated typical neoplastic features, in particular a prominent choline (trimethylamine -N+(CH3)3) resonance. The NMR choline resonance (which in vivo is almost certainly made up of several compounds containing a trimethylamine group) is considered to be related to cellular membrane turnover [31] and density and therefore might, in the future, be a useful marker of drug action targeted against cell growth. Clearly using an endocavitary probe (we used a pelvic phased array coil) or increasing magnetic field strength will improve the sensitivity of future MRS studies.



View larger version (70K):
[in this window]
[in a new window]
  		Figure 8. A short (echo time (TE)=20 ms) (top) and long (TE=135 ms) (bottom) single voxel 1H-MR spectrum (repetition time (TR)=1500 ms in both cases) from a locally advanced rectal cancer. The principal resonances are labelled and include "choline" (due to trimethylamine protons -N+(CH3)3) a marker of cell membrane turnover and the methyl (-CH2-) and methylene (-CH3) resonances of saturated lipid. The long T2 of the "choline" species compared with that of the lipids explains the relatively greater persistence of the "choline" peak over that of lipids in the long echo time spectrum (bottom left). The three orthogonal T1 localizer images are shown on the right and demonstrate the voxel sited within the confines of the tumour. Reproduced with permission, John Wiley & Sons, Inc.

 
Summary and future directions of multifunctional NMR studies
The study carried out on locally advanced rectal cancers at our institution has demonstrated it is possible to conduct a multifunctional study within a time acceptable to patients. Following initial refinements in the protocol, studies took no more than 50–60 min to complete. I would suggest that the principal value of this study has been to identify several potentially useful clinical associations. There are however several issues which require further study with the design and interpretation of the experiments. No reproducibility module was incorporated in the study and therefore there is no indication of the normal variation in the acquired data. This is of significance since it has become apparent recently that significant physiological variation can occur in pharmacokinetically modelled data [32]. In addition, the biological validation of the data remains to be established. For example, it is assumed, that functional NMR imaging is reflecting underlying angiogenesis but validated evidence is only just becoming available [26]. In addition, a study contradicting our findings has recently been reported [33]. Studies are required now to assess the reproducibility and clinical value of these measurements and to correlate the non-invasive imaging findings with whole mount histology and molecular markers of angiogenesis.


   Developing non-invasive strategies for the study of biliary drug excretion
Top
 Abstract
 Introduction
 Multifunctional studies in...
 Developing non-invasive...
References
 
Historical background
The concept of disease as an imbalance between the four "humours": black and yellow bile, phlegm and blood was first developed by Hippocrates of Cos (circa 460–377 BC). The great physician Claudius Galen whose influence extended to the emperor Marcus Aurelius promoted the concept and Galen's treatises were still influential up to the time of the Renaissance. Bile has therefore featured prominently throughout the history of medicine. It is likely that Reignier de Graff (1642–1673), the individual considered to be the father of modern toxicology, performed the first scientific studies as we understand them, of the biliary excretion of exogenous compounds in bile. In his studies de Graff developed the use of exteriorised biliary fistula for the systematic collection of bile thereby demonstrating turpentine excretion in the bile of turpentine fed dogs. Somewhat closer to imaging, JJ Abel (1857–1938) of Johns Hopkins was involved in the development of iodinated aromatic compounds subsequently used as oral and intravenous cholecystographic agents.

Rationale for developing a non-invasive methodology to study biliary drug excretion
Our interest in the study of biliary drug excretion was prompted by an in vivo pharmacokinetic study carried out at our institution [34]. The study was aimed at observing the serum metabolism of 5-fluorouracil (5FU) by high performance liquid chromatography (HPLC) and the simultaneous in vivo demonstration of the catabolism of 5FU (Figure 9Go) in the liver by 19F-MRS in patients receiving protracted venous infusion (PVI) 5FU. Here again we were exploiting the sensitivity of NMR nuclei to differences in local magnetic field causing the 19F nucleus of different compounds to resonate at slightly different frequencies thereby assigning a unique identifier to compounds. It was observed that high plasma levels of 5FU correlated with high-recorded levels of hepatic 5FU. Surprisingly though high hepatic levels of 5FU were associated with high recorded values of {alpha}-fluoro-{beta}-alanine (FBAL) the principal hepatic catabolite of 5FU. In the presence of negative feedback it would be reasonable to expect low levels of 5FU in the presence of high-recorded levels of hepatic FBAL (see Figure 9Go). One interpretation of the findings was that the FBAL was undergoing enterohepatic recirculation (EHC). From an evolutionary standpoint the EHC [35] is the mechanism by which the emulsifying bile salts are retained in the body following their secretion into bile. Similar to the excretion of bile, many classes of drug and xenobiotic undergo EHC [36]. We therefore hypothesised that if FBAL were undergoing EHC then we should at least expect a detectable catabolite signal in the gallbladder (GB) of patients. Clearly this would be short of proof of EHC but we postulated that the absence of a signal from the GB would make EHC an unlikely mechanism. It also occurred to us that such a tool could be of great benefit in the study of biliary excretion given that almost every class of drug and xenobiotic (including polar and non-polar compounds and several metals) regularly undergo biliary excretion [37]. In addition, it should be recalled that currently all forms of in vivo biliary sampling are invasive [38]. In humans this precludes the large-scale serial studies often necessary in toxicological research.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 9. The catabolic pathway of 5-fluorouracil (5FU). The majority of this process occurs in the liver and the rate-limiting step is the conversion of 5FU to dihydrofluorouracil by the enzyme dihydropyrimidine dehydrogenase. The principal catabolite is {alpha}-fluoro-{beta}-alanine that is predominantly excreted in urine. The species carboxy-fluoro-beta alanine (carboxy-FBAL) is rarely encountered since its formation is highly pH dependent. It would normally form under acidic conditions, which is rarely found in the liver and almost never in bile.

 
With the appropriate use of magnetic gradients it is possible to localize a spectroscopic signal in space. We employed the strategy of applying three orthogonal gradients to localize signal to a grid of voxels. This is termed three-dimensional chemical shift imaging (CSI). The trade-off in obtaining spatial sensitivity is loss of signal to noise, as signal now has to be gathered from individual voxels in the grid. On the plus side, free induction decays (FIDs) form the basis of the CSI signal and other than some loss from the application of the localizer gradients is the "best available" NMR signal (a stimulated echo, for example, gives only 50% of the signal available from a FID). Since a CSI sequence is essentially a pulse and acquire experiment it is possible to pulse at a rapid rate without having to be concerned about refocusing gradients. We also postulated that the chemical species containing the 19F nucleus would have a short "biological" T1 and therefore we could rapidly pulse with a 90° flip angle without excessive signal loss due to saturation.

To test our hypothesis we examined patients receiving protracted venous infusion of 5FU as part of an adjuvant chemotherapy protocol for rectal cancer. If 5FU or its catabolites were undergoing EHC then patients on PVI should, after initial dosing, arrive at a state of equilibrium where a constant amount of drug/catabolite is undergoing recirculation. To maximize our chances of acquiring a MRS signal, patients were asked to fast for a minimum of 4 h prior to MR examination in order for the GB to fill. The result was that a GB was typically contained within an individual voxel. We demonstrated that in these patients the exclusive 19F-MRS signal was localized to the gallbladder [39] (Figure 10Go) in an examination that lasted 40 min and included anatomical imaging and shimming. Perhaps unsurprisingly no 19F-MRS signal was recorded from a patient who had had a cholecystectomy in the past.



View larger version (167K):
[in this window]
[in a new window]
  		Figure 10. Exclusive 19F resonance arising from the gallbladder of an individual receiving protracted venous infusion of 5-fluorouracil (5FU). This example demonstrates a half Fourier single-shot turbo spin echo (HASTE) image of the upper abdomen with the 19F chemical shift imaging (CSI) grid overlain. The high signal intensity structure in the middle left grid is the gallbladder. In CSI, signals are acquired from each of the grids. Note the absence of a 19F resonance from the liver. The methodological set-up allowed the accurate determination of the resonant frequency of a particular chemical 19F species. We now believe the gallbladder resonance is due to bile acid conjugated {alpha}-fluoro-{beta}-alanine. Reproduced with permission, John Wiley & Sons, Inc [39].

 
Although a 19F signal had been detected from the GB, the chemical species required identification. If 5FU catabolites were undergoing EHC then it was likely to be in the form of bile acid conjugates of FBAL. Bile acid conjugates had previously been identified in the bile of two patients with external drainage of the common bile duct. In addition that research group synthesised several of the putative conjugates de novo thereby allowing confirmation of the expected chemical shifts and 19F J-couplings of the bile acid conjugates [40]. J-coupling (also referred to as scalar coupling) results from the interaction of two nuclear spins on the same molecule through distortions in their electron clouds. J-coupling is therefore a phenomenon that occurs "through bonds" rather than "through space" and is the cause of a split resonance best appreciated in high-field (analytical) scanning. It was the similar physicochemical properties of taurine including pK and ionicity to FBAL that led to the proposal that FBAL might be competing with endogenous taurine for conjugation to primary and secondary bile acids (the amino acids taurine and glycine principally conjugate bile acids in man). In addition the EHC of bile acid conjugated FBAL had previously been demonstrated in an animal model using conventional biliary and enteric sampling techniques [41]. There was therefore considerable evidence to support that the resonance we had detected in the GB was from bile acid conjugated FBAL. We had therefore purposely designed our MRS protocol to accurately reference chemical shifts. We found that the exclusive 19F signal detected from the GB resonated approximately 2.2 ppm downfield of the expected resonance of simple FBAL. This was exactly at the expected site of resonance for FBAL-conjugated bile acids. Since none of the other known metabolites of 5FU [42] resonated at that point we surmised that the resonance was due to FBAL-conjugated bile acids. Further definition of which bile acids might be conjugating with FBAL was impossible since accurate resolution of the 19F J-couplings was impossible at 1.5 T. In addition, several of the bile acid conjugates resonate at the same chemical shift thereby confounding resolution.

A further opportunity arose to support the biliary nature of the 19F signal in a group of patients receiving bolus intravenous 5FU. Although there would not be time to arrive at an equilibrium state as postulated for the PVI patients the large doses administered in a bolus regimen gave us some hope that we might sequentially observe a hepatic and then biliary signal. In a separate group of patients we followed 5FU catabolite kinetics using on this occasion a single voxel technique in order to maximize signal from either the gallbladder or liver. As expected, a hepatic FBAL signal was detected within 20 min of systemic 5FU administration. At that time no 19F-MRS GB signal was detected. At 1 h though a clear 19F-MRS signal was recorded from the GB (Figure 11Go). The GB signal was once again approximately 2.2 ppm downfield of the hepatic signal. Evidence therefore of a different 19F compound in the GB than that detected in the liver and together with previous literature evidence is strongly suggestive of the presence of FBAL-bile acid conjugates.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 11. (a) Unlocalized and (b and c) single voxel 19F experiments in a patient receiving a bolus dose of 5-fluorouracil (5FU). The two black arrows indicate the expected resonance of {alpha}-fluoro-{beta}-alanine (right arrow) and bile acid conjugated {alpha}-fluoro-{beta}-alanine (left arrow). Note how both resonances are visible in the (a) unlocalized sequence. As expected, {alpha}-fluoro-{beta}-alanine is detected in the liver (b) and note the single resonance detected in the gallbladder centred on the expected resonance of bile acid conjugated {alpha}-fluoro-{beta}-alanine.

 
Biliary studies: future perspectives
Developing strategies dealing particularly with the effects of physiological motion together with the use of increasing field strengths for in vivo studies promises, I believe, increased potential for non-invasive studies of this physiologically and pharmacologically important compartment. Although it is not within the scope of this review we have now demonstrated the biliary excretion of drugs and xenobiotics containing the 19F, 1H or 31P nucleus [43], a true multinuclear approach. In respect of the FBAL-bile acid conjugate work, we have demonstrated and confirmed the presence of FBAL-bile acid conjugate in an animal model [44]. This will allow studies into the therapeutic consequences of modulating the EHC of FBAL. There is in addition no need to exclusively study exogenous compounds as in the 5FU studies. There is no reason why the endogenous composition of human GB bile in vivo cannot be studied in situ [45], allowing perhaps the non-invasive study of cholelithiasis.

Summary and future directions
In 1979, Sir Godfrey speculated on the use of NMR to probe the biophysical characteristics of tissues and organisms. Incorporating developments in hardware and computer technology we have begun to non-invasively interrogate biological processes in vivo. As our knowledge progresses so new challenges and questions will arise. Particularly in human studies there is a great need to validate the results of studies such as ours with the underlying biology. It is also absolutely necessary to maintain high standards of scientific methodology and to be rigorous in assessing the reproducibility of our techniques.

It is little over a 150 years ago that Gregor Mendel discovered the scientific basis that now support Charles Darwin's view of nature. Just over a hundred years ago William Roentgen discovered X-rays whilst it is 50 years since Crick and Watson elucidated the crystal structure of DNA. 30 years ago Sir Godfrey and his colleagues revolutionized medical imaging through the development of CT-scanning. We are surely at the brink of a new "biological" era through the sequencing of several animal genomes including that of humans. Medicine, industry and business have their sights set on molecular targets. It will be the role of imaging scientists and clinicians to bring those processes to light. The new challenge of functional/molecular imaging [46–51] will require us to build on our previous studies but also to focus on the clinically and biologically relevant questions.

Ethics
All human and animal experiments undertaken by the author and referred to in this text received the appropriate institutional ethical approval.


   Acknowledgments
 
I am grateful to Mr David Collins, Professor Janet Husband and Professor Martin Leach for their critical review of the article. I also wish to acknowledge Dr IJ Rowland who supervised my PhD that included some of the biliary work described in this text.


   Footnotes
 
The research work described here was generously supported by Cancer Research UK grant SP1780/0103 and a research bursary from the British Journal of Surgery (awarded to Mr M George). Back

Received for publication June 20, 2003. Revision received September 17, 2003. Accepted for publication September 29, 2003.


   References
Top
 Abstract
 Introduction
 Multifunctional studies in...
 Developing non-invasive...
 References
 

  1. Husband JE, Gwyther SJ, Rankin S. Monitoring tumor response. Abdom Imaging 1999;24:618–21.[CrossRef][Medline]
  2. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.
  3. Gillies RJ, Raghunand N, Karczmar GS, Bhujwalla ZM. MRI of the tumor microenvironment. J Magn Reson Imaging 2002;16:430–50.[CrossRef][Medline]
  4. Ross BD, Zhao YJ, Neal ER, Stegman LD, Ercolani M, Ben-Yoseph O, et al. Contributions of cell kill and posttreatment tumor growth rates to the repopulation of intracerebral 9L tumors after chemotherapy: an MRI study. Proc Natl Acad Sci USA 1998;95:7012–7.[Abstract/Free Full Text]
  5. Brown G, Richards CJ, Newcombe RG, Dallimore NS, Radcliffe AG, Carey DP, et al. Rectal carcinoma: thin-section MR imaging for staging in 28 patients. Radiology 1999;211:215–22.[Abstract/Free Full Text]
  6. Tofts PS, Brix G, Buckley DL, Evelhoch JL, Henderson E, Knopp MV, et al. Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–32.[CrossRef][Medline]
  7. Baustert IC, Padhani AP, Revell P, Leach MO. From qualitative to quantitative measurements of permeability, leakage space and relative blood volume. Proc Intl Soc Magn Reson Med 1998;6:1655.
  8. Miller AB, Hoogstraten B, Staquet M. Reporting results of cancer treatment. Cancer 1981;47:207–14.[CrossRef][Medline]
  9. Derrick TS, McCord EF, Larive CK. Analysis of protein/ligand interactions with NMR diffusion measurements: the importance of eliminating the protein background. J Magn Reson 2002;155:217–25.[CrossRef][Medline]
  10. Wheeler-Kingshott CA, Thomas DL, Lythgoe MF, Guilfoyle D, Williams SR, Doran SJ. Burst excitation for quantitative diffusion imaging with multiple b-values. Magn Reson Med 2000;44:737–45.[CrossRef][Medline]
  11. Dzik-Jurasz A, Domenig C, George M, Wolber J, Padhani A, Brown G, et al. Diffusion MRI for prediction of response of rectal cancer to chemoradiation. Lancet 2002;360:307–8.[CrossRef][Medline]
  12. Lemaire L, Howe FA, Rodrigues LM, Griffiths JR. Assessment of induced rat mammary tumour response to chemotherapy using the apparent diffusion coefficient of tissue water as determined by diffusion-weighted 1H-NMR spectroscopy in vivo. Magma 1999;8:20–6.
  13. Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and necrotic fraction in human melanoma xenografts by diffusion weighted magnetic resonance imaging. Magn Reson Med 2000;43:828–36.[CrossRef][Medline]
  14. Horwich A. Cancer Research Campaign review of radiobiology research. Br J Cancer 1993;67:198–201.[Medline]
  15. Galons JP, Altbach MI, Paine-Murrieta GD, Taylor CW, Gillies RJ. Early increases in breast tumor xenograft water mobility in response to paclitaxel therapy detected by non-invasive diffusion magnetic resonance imaging. Neoplasia 1999;1:113–7.[CrossRef][Medline]
  16. Chinnaiyan AM, Prasad U, Shankar S, Hamstra DA, Shanaiah M, Chenevert TL, et al. Combined effect of tumor necrosis factor-related apoptosis-inducing ligand and ionizing radiation in breast cancer therapy. Proc Natl Acad Sci USA 2000;97:1754–9.[Abstract/Free Full Text]
  17. Jennings D, Hatton BN, Guo J, Galons JP, Trouard TP, Raghunand N, et al. Early response of prostate carcinoma xenografts to docetaxel chemotherapy monitored with diffusion MRI. Neoplasia 2002;4:255–62.[CrossRef][Medline]
  18. Zhao M, Pipe JG, Bonnett J, Evelhoch JL. Early detection of treatment response by diffusion-weighted 1H-NMR spectroscopy in a murine tumour in vivo. Br J Cancer 1996;73:61–4.[Medline]
  19. Stegman LD, Rehemtulla A, Hamstra DA, Rice DJ, Jonas SJ, Stout KL, et al. Diffusion MRI detects early events in the response of a glioma model to the yeast cytosine deaminase gene therapy strategy. Gene Ther 2000;7:1005–10.[CrossRef][Medline]
  20. Poptani H, Puumalainen AM, Grohn OH, Loimas S, Kainulainen R, Yla-Herttuala S, et al. Monitoring thymidine kinase and ganciclovir-induced changes in rat malignant glioma in vivo by nuclear magnetic resonance imaging. Cancer Gene Ther 1998;5:101–9.[Medline]
  21. Chenevert TL, Stegman LD, Taylor JM, Robertson PL, Greenberg HS, Rehemtulla A, et al. Diffusion magnetic resonance imaging: an early surrogate marker of therapeutic efficacy in brain tumors. J Natl Cancer Inst 2000;92:2029–36.[Abstract/Free Full Text]
  22. Mardor Y, Roth Y, Lidar Z, Jonas T, Pfeffer R, Maier SE, et al. Monitoring response to convection-enhanced taxol delivery in brain tumor patients using diffusion-weighted magnetic resonance imaging. Cancer Res 2001;61:4971–3.[Abstract/Free Full Text]
  23. Bikfalvi A, Bicknell R. Recent advances in angiogenesis, anti-angiogenesis and vascular targeting. Trends Pharmacol Sci 2002;23:576–82.[CrossRef][Medline]
  24. George ML, Dzik-Jurasz AS, Padhani AR, Brown G, Tait DM, Eccles SA, et al. Non-invasive methods of assessing angiogenesis and their value in predicting response to treatment in colorectal cancer. Br J Surg 2001;88:1628–36.[CrossRef][Medline]
  25. Midgley R, Kerr D. Colorectal cancer. Lancet 1999;353:391–9.[CrossRef][Medline]
  26. Padhani AR. Dynamic contrast-enhanced MRI in clinical oncology: current status and future directions. J Magn Reson Imaging 2002;16:407–22.[CrossRef][Medline]
  27. Smith ICP, Stewart LC. Magnetic resonance spectroscopy in medicine: clinical impact. Prog Nucl Mag Res Sp 2002;40:1–34.
  28. Leach MO. Introduction to in vivo MRS of cancer: new perspectives and open problems. Anticancer Res 1996;16:1503–14.[Medline]
  29. Findlay MP, Leach MO. In vivo monitoring of fluoropyrimidine metabolites: magnetic resonance spectroscopy in the evaluation of 5-fluorouracil. Anticancer Drugs 1994;5:260–80.[Medline]
  30. Dzik-Jurasz AS, Murphy PS, George M, Prock T, Collins DJ, Swift I, et al. Human rectal adenocarcinoma: demonstration of 1H-MR spectra in vivo at 1.5 T. Magn Reson Med 2002;47:809–11.[CrossRef][Medline]
  31. Miller BL, Chang L, Booth R, Ernst T, Cornford M, Nikas D, et al. In vivo 1H MRS choline: correlation with in vitro chemistry/histology. Life Sci 1996;58:1929–35.[CrossRef][Medline]
  32. Galbraith SM, Lodge MA, Taylor NJ, Rustin GJ, Bentzen S, Stirling JJ, et al. Reproducibility of dynamic contrast-enhanced MRI in human muscle and tumours: comparison of quantitative and semi-quantitative analysis. NMR Biomed 2002;15:132–42.[CrossRef][Medline]
  33. de Vries A, Griebel J, Kremser C, Judmaier W, Gneiting T, Debbage P, et al. Monitoring of tumor microcirculation during fractionated radiation therapy in patients with rectal carcinoma: preliminary results and implications for therapy. Radiology 2000;217:385–91.[Abstract/Free Full Text]
  34. Findlay MP, Raynaud F, Cunningham D, Iveson A, Collins DJ, Leach MO. Measurement of plasma 5-fluorouracil by high-performance liquid chromatography with comparison of results to tissue drug levels observed using in vivo 19F magnetic resonance spectroscopy in patients on a protracted venous infusion with or without interferon-alpha. Ann Oncol 1996;7:47–53.[Abstract/Free Full Text]
  35. Carey MC, Duane WC. Enterohepatic circulation. In: Arias IM, editor. The liver: biology and pathobiology. 3rd edn. New York: Raven Press, 1994;719–67.
  36. Klaassen CD, Watkins JB, 3rd. Mechanisms of bile formation, hepatic uptake, and biliary excretion. Pharmacol Rev 1984;36:1–67.[Medline]
  37. Siegers CP, Watkins III JB, editors. Biliary excretion of drugs and other chemicals. 1st edn. New York: Gustav Fischer Verlag, 1991.
  38. Schofl R. Diagnostic endoscopic retrograde cholangiopancreatography. Endoscopy 2001;33:147–57.[CrossRef][Medline]
  39. Dzik-Jurasz AS, Collins DJ, Leach MO, Rowland IJ. Gallbladder localization of 19F MRS catabolite signals in patients receiving bolus and protracted venous infusional 5-fluorouracil. Magn Reson Med 2000;44:516–20.[CrossRef][Medline]
  40. Malet-Martino MC, Bernadou J, Martino R, Armand JP. 19F NMR spectrometry evidence for bile acid conjugates of alpha-fluoro-beta-alanine as the main biliary metabolites of antineoplastic fluoropyrimidines in humans. Drug Metab Dispos 1988;16:78–84.[Abstract]
  41. Zhang RW, Barnes S, Diasio RB. Disposition and metabolism of 2-fluoro-beta-alanine conjugates of bile acids following secretion into bile. Biochim Biophys Acta 1991;1096:179–86.[Medline]
  42. Heggie GD, Sommadossi JP, Cross DS, Huster WJ, Diasio RB. Clinical pharmacokinetics of 5-fluorouracil and its metabolites in plasma, urine, and bile. Cancer Res 1987;47:2203–6.[Abstract/Free Full Text]
  43. Payne GS, Dzik-Jurasz ASK, Mancini L, Vaidya S, Leach MO. Double-resonance and bio-distribution studies of ifosfamide in vivo with 31P MRS. Proc Intl Soc Magn Reson Med 2002;10:2144
  44. Dzik-Jurasz AS. Development and application of MR strategies for the in vivo study of the tumour and bile microenvironment in anticancer research [PhD]. University of London: London, 2001.
  45. Prescot A, Collins DJ, Leach MO, Dzik-Jurasz AS. Human gallbladder bile: noninvasive investigation in vivo with single-voxel 1H MR spectroscopy. Radiology 2003; 229:587–92.[Abstract/Free Full Text]
  46. Hoffman JM, Menkens AE. Molecular imaging in cancer: future directions and goals of the National Cancer Institute. Acad Radiol 2000;7:905–7.[CrossRef][Medline]
  47. Clarke LP, Croft BY, Menkens A, Torres-Anjel MJ, Hoffman JM, Sullivan DC. National Cancer Institute initiative for development of novel imaging technologies. Acad Radiol 2000;7:481–3.[CrossRef][Medline]
  48. Sullivan DC. Biomedical imaging symposium: visualizing the future of biology and medicine. Radiology 2000;215:634–8.[Free Full Text]
  49. Hillman BJ, Neiman HL. Translating molecular imaging research into radiologic practice: summary of the proceedings of the American College of Radiology Colloquium, April 22–24, 2001. Radiology 2002;222:19–24.[Abstract/Free Full Text]
  50. Li KC. Biomedical imaging in the postgenomic era: opportunities and challenges. Acad Radiol 2002;9:999–1003.[CrossRef][Medline]
  51. Weissleder R, Mahmood U. Molecular imaging. Radiology 2001;219:316–33.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Br. J. Radiol.Home page
A Jackson
Imaging microvascular structure with contrast enhanced MRI
Br. J. Radiol., December 1, 2003; 76(suppl_2): S159 - S173.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dzik-Jurasz, A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dzik-Jurasz, A

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS