British Journal of Radiology (2003) 76, S60-S80
© 2003 British Institute of Radiology
doi: 10.1259/bjr/15334380
MRI for assessing antivascular cancer treatments
A R Padhani, MB BS, FRCP, FRCR
Paul Strickland Scanner Centre, Mount Vernon Hospital, Rickmansworth Road, Northwood, Middlesex HA6 2RN, UK
 |
Abstract
|
|---|
Selective antiangiogenesis and vascular targeting drugs hold out the promise of improved efficacy and tolerability for anticancer treatments. Early phase 1 drug trials have shown good tolerability for antiangiogenesis agents with biological activity below the maximum tolerated dose. Advanced clinical trials have demonstrated that morphological assessments of tumour response are of limited value in gauging the efficacy of treatment. MRI is a versatile technique which is sensitive to contrast mechanisms that can be affected by antivascular treatments; this use for MRI has been validated in xenografts and humans. Dynamic contrast-enhanced MRI (DCE-MRI), which demonstrates tissue perfusion and permeability, is being used clinically as a pharmacodynamic indicator of biological activity for antivascular cancer drugs. Early data show that DCE-MRI studies can define the biologically active dose and predict the efficacy of treatment on the basis of changes observed. MRI with macromolecular contrast media (MMCM) depicts microvessel permeability and fractional plasma volume. Xenograft studies with MMCM have shown great promise for evaluating antivascular treatments but this has not been used clinically. Intrinsic susceptibility-weighted MRI, which is sensitive to blood oxygenation and flow, is emerging as a technique that may be able to monitor vascular targeting therapies.
|
Introduction
|
|---|
Imaging assessments of the functional tumour vasculature could have widespread clinical applications; recently, developments in angiogenesis imaging have gained greater impetus by the development of antivascular drugs that target the functioning tumour microvasculature. The need for imaging biomarkers that inform on drug action non-invasively has been widely recognised. Several imaging techniques are able to assess human tumours with respect to their angiogenic status. All the major imaging techniques are discussed in this special "angiogenesis imaging" issue of the British Journal of Radiology. MRI is used experimentally and clinically to characterize microvasculature, providing information about tumour microvessel structure and function [1]. MRI techniques are sensitive to ultrastructural and functional abnormalities that are characteristic of malignant vasculature. Thus, MRI can depict: (1) spatial heterogeneity of perfusion, which reflects tissue vascular density where areas of low vascular density mix with regions of high angiogenic activity [2]; (2) hyperpermeability to macromolecules due to poorly formed, fragile vessels which have large endothelial cell gaps or fenestrae [3], incomplete basement membrane and which lack complete pericyte or smooth muscle associations with endothelial cells [4]; (3) increased vascular volume and flow, and high vascular tortuosity; and (4) tissue hypoxia.
MRI techniques can be divided into extrinsic (contrast media enhanced) and intrinsic (non-enhanced) methods [5]. The former can be further divided by the type of contrast medium utilized; (i) low molecular weight (MW) agents (<1 kDa) that rapidly diffuse in the extracellular fluid space (ECF agents), (ii) intermediate (MW 10–30 kDa) and high molecular agents (MW >30 kDa) designed for prolonged intravascular retention (macromolecular contrast media (MMCM) or blood pool agents) [6], and (iii) agents that accumulate at sites of concentrated angiogenesis mediating molecules [7]. This review concentrates on non-invasive characterization of tumour neovasculature with dynamic contrast-enhanced MRI (DCE-MRI) using low-molecular weight contrast agents and explains how perfusion and permeability data can be extracted depending on the technique utilized [8–10]. The potential of DCE-MRI to demonstrate the effects of antivascular cancer treatments will be shown. DCE-MRI using MMCM has been validated in pre-clinical studies as being an appropriate biomarker for monitoring the effects of antiangiogenic drugs and although no human studies have been published in this regard, their potential role will be reviewed [6]. Tumour angiogenesis can also be analysed using intrinsic susceptibility weighted or blood oxygenation level dependent (BOLD) MRI [11]. BOLD imaging can be used for mapping mature and immature vessels and their differential sensitivity to perturbations in vascular endothelial growth factor (VEGF) expression [12] and for monitoring the effects of vascular targeting agents [13, 14]. These applications of BOLD MRI will be discussed. Other MRI techniques that are able to inform on tissue blood flow such as flow related enhancement and diffusion weighted imaging will not be considered but interested readers are invited to review the recent article by Neeman and Dafni for an excellent overview [1].
|
MRI with low molecular weight contrast media
|
|---|
Contrast agent kinetics
DCE-MRI is able to distinguish malignant from benign and normal tissues by exploiting differences in contrast agent behaviour in their respective microcirculations. When a bolus of paramagnetic, low molecular weight contrast agent passes through a capillary bed, it is transiently confined within the vascular space. This "first pass" includes the arrival of contrast medium and lasts for a few cardiac cycles. In most tissues except the brain, testes and retina, the contrast agent rapidly passes into the extravascular–extracellular space (EES, also called leakage space – ve) at a rate determined by the permeability of the microvessels, their surface area and by blood flow (Figure 1
). In tumours, typically 12–45% of the contrast media leaks into the EES during the first pass [15]. The transfer constant (Ktrans) describes the transendothelial transport of low molecular weight contrast medium from the vascular to the interstitial space. Three major factors determine the behaviour of low molecular weight contrast media in tissues during the first few minutes after injection; blood perfusion, transport of contrast agent across vessel walls and diffusion of contrast medium in the interstitial space. If the delivery of the contrast medium to a tissue is insufficient (flow-limited situations or where vascular permeability is greater than inflow) then blood perfusion will be the dominant factor determining tissue enhancement and Ktrans approximates to tissue blood flow per unit volume [16], this condition is commonly found in extracranial tumours due to high microvessel permeability. If tissue perfusion is sufficient and transport out of the vasculature does not deplete intravascular contrast medium concentration (non-flow limited or permeability limited) then transport across the vessel wall is the major factor that determines tissue enhancement (Ktrans then approximates to permeability surface area product – PS). The latter circumstance occurs in areas of radiation fibrosis, in the presence of an intact or partially intact blood–brain barrier but can also occur in extracranial tumours usually after treatment.
As low molecular weight contrast media do not cross cell membranes, the volume of distribution is effectively the EES (ve). Contrast medium also begins to diffuse into tissue compartments further removed from the vasculature including areas of necrosis and fibrosis. Over a period typically lasting several minutes to an hour, the contrast agent diffuses back into the vasculature (described by the rate constant or kep) from where it is excreted (usually by the kidneys although some ECF contrast media have significant hepatic excretion). When capillary permeability is very high, the return of contrast medium is typically rapid resulting in faster washout as plasma contrast medium concentrations fall. Contrast medium elimination from very slow-exchange tissues such as fibrosis and necrosis occurs slowly, explaining the persistent delayed enhancement described in some tumours such as cholangiocarcinoma and hepatic colorectal metastases.
MRI sequences can be designed to be sensitive to the vascular phase of contrast medium delivery (so-called T2* methods which reflect on tissue perfusion and blood volume) [17, 18]. T1 weighted sequences are sensitive to the presence of contrast medium in the EES and thus reflect microvessel perfusion, permeability and extracellular leakage space. These two DCE-MRI methods are compared in Table 1.
|
T2* weighted DCE-MRI
|
|---|
Data acquisition
Perfusion-weighted images can be obtained with "bolus-tracking techniques" that monitor the passage of contrast material through a capillary bed [17, 18]. Within the vascular space and in the immediate vicinity, paramagnetic contrast media produce magnetic field (Bo) changes that result in a decrease in the relaxation times of surrounding tissues (Figures 2 and 3). Susceptibility-weighted (T2* weighted) echo-planar spin-echo sequences are more sensitive to capillary blood flow compared with echo planar gradient-echo sequences, which also incorporate signals from larger vessels [19]. The degree of signal intensity loss is dependent on the vascular concentration of the contrast agent and microvessel size [20] and density. The signal-to-noise ratio (SNR) of T2* weighted DCE-MR images can be improved by using high doses of contrast medium (i.e. 
ge;0.2 mmol kg–1 body weight) [21]. Standard spoiled gradient-echo sequences on conventional MRI systems can characterize these effects but are limited to a few slices. High specification, echo-planar enabled MRI systems capable of rapid image acquisition allow greater anatomical coverage. However, echo-planar sequences have limited applications in extracranial tissues due to great intrinsic sensitivity to susceptibility-inducing environments (e.g. highly concentrated contrast media and bowel gas/tissue boundaries) which can result in spatial misregistration of major vessels during the first passage of the contrast agent thorough the vessels [22].
Quantification
Tracer kinetic principles can be used to provide estimates of relative blood volume (rBV), relative blood flow (rBF) and mean transit time (MTT) derived from the first-pass of contrast agent through the microcirculation [17, 18, 23] (Figure 3
). MTT is the average time the contrast agent takes to pass through the tissue being studied. These variables are related by the central volume theorem equation (BF=BV/MTT). A number of conditions of the central volume theorem cannot be met in biological tissues. For example, injection time is not instantaneous and as the arterial input function is not typically measured; these parameter estimates are usually qualitative or "relative". The most robust parameter which can be extracted reliably from first pass techniques is rBV, which is obtained from the integral of the time series data during the first pass of the contrast agent [24]. This cannot readily be done for extracranial tumours because of the loss of compartmentalization of the contrast medium (see below for further details). Instead the time series data are fitted to a gamma-variate function from which the parameters rBV, MTT and rBF are derived. An additional parameter that can be derived from the T2* DCE-MRI data is the tortuosity index, which is the difference between the total time series integral and the integral of the gamma variate derived from the first pass [25]. The tortuosity index reflects the abnormal retention of contrast material due to anatomical abnormalities of the tumour vasculature described above. The tortuosity index can only be derived for brain tumours because there is no or little loss of compartmentalization of contrast medium bolus during the first pass.
Absolute quantification of T2* weighted kinetic parameters can be obtained by measuring the changing concentration of contrast agent in the feeding vessel, and in this way, quantified perfusion parameters in normal brain and of low grade gliomas have been obtained [26, 27]. Absolute quantification is not currently possible for evaluation of visceral tissues and tumours due to a number of limitations discussed below. From a practical perspective, it is not always necessary to quantify T2* weighted DCE-MRI data to obtain insights of the spatial distribution of tissue perfusion. Simple subtraction images can demonstrate the maximal signal attenuation, a semiquantitative parameter which has been strongly correlated with relative blood flow and volume in tumours [28, 29]. Subtraction analysis should only be done if the is a linear relationship between rBV and rBF; that is, when MTT is in a narrow range (Figure 2). In non-necrotic tumours the MTT is often in a narrow range, which is in marked contrast to the situation in ischaemia-induced cerebral stroke where significant lengthening of brain MTT is a characteristic feature (unchanged rBV but reduced rBF).
Limitations
Physiological effects that hinder measurements of perfusion in tumours include non-laminar flow (which arises from the presence of irregular calibre vessels), non-dichotomous branching and high vascular permeability and variations in the haematocrit fraction as blood passes through a vascular bed. In addition, factors such as machine stability, patient motion and intrinsic patient variables, particularly cardiac output and upstream stenoses, can affect computations. Re-circulation and marked contrast leakage into the extracellular space during the first pass of contrast medium are the principle causes resulting in falsely low blood volume values. Extracranial tumours have very leaky blood vessels and the loss of contrast medium compartmentalization is observed by the failure of the signal intensity to return to baseline (Figure 2). Furthermore, the T1 signal enhancing effects of contrast medium leaking from blood vessels can counteract T2* signal lowering effects. Quantitative imaging is thus most reliably used for normal brain and non-enhancing brain lesions because the contrast medium is completely or largely retained within the intravascular space. Solutions to overcoming these problems include the use of non-gadolinium susceptibility contrast agents based on the element dysprosium or ultrasmall superparamagnetic iron oxide particles (USPIOs), which have strong T2* effects but weak T1 effects [30, 31]. Preliminary results indicate that dysprosium-based relative cerebral blood volume (rCBV) maps are superior to those obtained with gadolinium chelates [32]. USPIOs designed for bolus injection have the advantage of being retained within the vascular space during the first pass due to their high molecular weight [33, 34]. Encouraging early clinical results using USPIOs are beginning to appear in the literature [31, 35]. Solutions for counteracting the T1 enhancing effects of gadolinium chelates include idealized model fitting (gamma variate function), pre-dosing with contrast medium to saturate the leakage space and by using dual or multiecho imaging sequences that minimize T1 sensitivity [36]. We favour the latter techniques and illustrative images of computed rBV, rBF and MTT for a breast cancer are shown in Figure 3.
Clinical experience
Quantitative imaging is currently most reliable for normal brain and non-enhancing brain lesions because the contrast medium is retained within the intravascular space. T2* weighted perfusion mapping techniques have progressively entered neurological practice [29, 37, 38]. Clinical applications include characterization of tumour vascularity [39–42], follow-up of treatment response [27, 29, 43, 44] and the study of stroke [45]. Areas of high tumour rCBV are readily visible in patients with brain gliomas [41, 46] and appear to correlate with mitotic activity (information on tumour grade) and vascularity assessed angiographically but not with cellular atypia, endothelial proliferation, necrosis or cellularity [41]. There is a paucity of data correlating T2* kinetic parameters with immunohistochemical microvessel density (MVD) in human brain tumours [47, 48]. rCBV maps appear to have a high negative predictive value in excluding the presence of high grade tumour in untreated patients regardless of their enhancement characteristics on T1 weighted MRI. In low grade gliomas, homogeneous low rCBV is found whereas higher grade tumours display both low and high rCBV components [49]. rCBV can thus be used to direct stereotactic biopsy [50, 51].
There is very little literature data on the usage of T2* weighted DCE-MRI outside the brain. Qualitative observations of signal loss observed on T2* weighted sequences after injection of gadolinium-containing contrast agents have been reported in preliminary clinical studies to characterize liver, breast and brain tumours. For example, Ichikawa et al [52] were able to discriminate between liver metastases, haemangiomas and hepatomas on the basis of characteristic signal intensity changes on echo-planar MRI. Both Kuhl et al [53] and Kvistad et al [54] have qualitatively evaluated the value of T2* weighted DCE-MRI for characterizing breast lesions. Both studies showed strong decreases in signal intensity (equivalent to high rBV/rBF) in malignant tissues whereas susceptibility effects in fibroadenomas were minor (equivalent to low rBV/rBF). The latter studies showed that it was possible to differentiate carcinomas from fibroadenomas with high specificity using T2* characteristics despite significant overlap in T1 enhancement patterns. The pathophysiological explanation for these observations probably relate to differences in microvessel arrangements, density and size in malignant tumours and fibroadenomas [55]. Quantitative T2* weighted DCE-MRI have been used to monitor the effects of chemotherapy in breast cancer. Ah-See et al [56] have observed that rBV and rBF were as effective as T1 weighted kinetic parameters in predicting non-responsiveness to neoadjuvant chemotherapy.
Recently, T2* weighted MRI has been used to monitor the antivascular treatment effects of thalidomide in combination with carboplatin in patients with recurrent malignant gliomas [57]. This preliminary report showed that changes on T2* weighted images correlated more closely with the clinical status of patients than contrast enhanced T1 weighted images, possibly because enhancement on T1 weighted imaging is not specific for active disease. T2* weighted DCE-MRI has also been applied to study the effects of the anti-VEGF tyrosine kinase inhibitor PTK/ZK in brain gliomas [58] but studies to evaluate antivascular drugs in visceral tumours have not yet been reported.
|
T1 weighted DCE-MRI
|
|---|
Data acquisition
Extracellular contrast media readily diffuse from the blood into the EES of extracranial tissues at a rate determined by tissue perfusion, permeability of the capillaries and their surface area. Shortening of the T1 relaxation time caused by the contrast medium is the mechanism of tissue enhancement. Most DCE-MRI studies employ T1-weighted gradient-echo, saturation recovery/inversion recovery snapshot sequences (e.g. turboFLASH) or echo-planar sequences [59–61] (Figure 4). The choice of sequence and parameters used is dependent on intrinsic advantages and disadvantages of the sequences, taking into account T1 sensitivity, anatomical coverage, acquisition times, susceptibility to artefacts arising from magnetic field inhomogeneities and accuracy for quantification.
The pattern of signal intensity enhancement seen on T1 weighted DCE-MRI is dependent on a number of physical and physiological factors. Physical factors include the native T1 relaxation rate of the tissue, dose of contrast agent, imaging sequence and parameters used and on machine gain and scaling factors. Physiological factors include tissue perfusion, capillary surface area and permeability to contrast agent, and volume of the extracellular leakage space.
Figure 4 shows that T1 weighted kinetic enhancement curves have three distinct phases; the upslope, maximum enhancement and washout. It is generally recognised that the upslope gradient in tumours is highly dependent on tissue perfusion and permeability with perfusion predominating principally due to high blood volume and high first pass extraction. Maximum enhancement is related to the total uptake concentration of the contrast medium in the interstitial space and washout rate is associated with tissue contrast agent concentration decrease and is strongly related to vascular permeability. If it is assumed that tissue enhancement occurs due to contributions from vascular and extravascular compartments (see two-compartment modelling below) then it is possible to separate these inputs mathematically using deconvolution techniques [62] which is helpful for understanding the shape of kinetic curves [63]. The dominant contribution of perfusion to the upslope of T1 weighted DCE-MRI enhancement curves can be verified by correlating T1 and T2* weighted DCE-MRI enhancement curves and corresponding kinetic pixel maps (Figure 5).
Quantification
Signal enhancement seen on T1 weighted DCE-MRI can be assessed in two ways: by the analysis of signal intensity changes (semiquantitative) and/or by quantifying contrast agent concentration change (or 
R1) using pharmacokinetic modelling techniques. Semiquantitative parameters describe signal intensity changes using a number of descriptors. These parameters include curve shape [64, 65], onset time (time from injection or appearance in an artery to the arrival of contrast medium in the tissue of interest), gradient of the upslope of enhancement curves, maximum signal intensity and washout gradient. As the rate of enhancement is important for improving the specificity of diagnoses, parameters that include a timing element are often used (e.g. maximum intensity time ratio (MITR) [66] and maximum focal enhancement at 1 min) [67, 68]. The uptake integral or initial area under the signal intensity curve (IAUC) or contrast medium concentration curve (IAUGC) have also been studied [69]. Experimental data indicate that in practice IAUGC at 30 s correlates with transfer constant Ktrans whereas IAUGC at 90 s correlates with lesion leakage space (ve) in brain tumours [70]. Thus, semiquantitative parameters have a close but complex and not totally defined link to underlying tissue physiology and contrast agent kinetics. Semiquantitative parameters have the advantage of being relatively straightforward to calculate but have limitations. These limitations include the fact that they do not accurately reflect contrast medium concentration in tissues and can be influenced by scanner settings (including gain and scaling factors). These factors limit the usefulness of semiquantitative parameters and make between-patient and between-system comparisons difficult. Semiquantitative parameters are not recommended as biomarkers for the evaluation antivascular cancer trials (except for IAUGC), which should instead use quantitative kinetic parameters discussed below [71].
Quantitative techniques use pharmacokinetic modelling applied to changes in tissue contrast agent concentration or T1 relaxivity. In general, it is not recommended that pharmacokinetic modelling be done on signal intensity data unless it is has been shown that there is a direct relationship between signal intensity and contrast agent concentration over the entire range expected in tissues. Signal intensity values observed during dynamic acquisition can be used to estimate contrast agent concentration at each time point [61, 72]. Mathematically fitting these data to pharmacokinetic models yields quantitative kinetic parameters (Figure 6). A two compartment model modified from the Kety formula relating the change of tissue tracer concentration to the difference between arterial plasma and interstitial fluid concentrations [73] is most often used [59, 74, 75]. For a detailed discussion on pharmacokinetic modelling techniques, readers are directed to the review by Tofts [76]. Tofts et al published a consensus document that standardized the form and terms of this compartmental modelling approach for DCE-MRI. Examples of modelling parameters include the volume transfer constant of the contrast agent (Ktrans – formally called permeability–surface area product (PS) per unit volume of tissue – unit min–1), leakage space as a percentage of unit volume of tissue (ve – unit %) and the rate constant (kep; also called K21 – unit min–1) (Figure 1). These standard parameters are related mathematically (kep=Ktrans/ve) [16].
As noted under the description of contrast agent kinetics, tumours are markedly heterogeneous in their perfusion. There are areas where the PS is high compared with flow (F) and it is these tissues that are "flow limited". In these areas, Ktrans estimates are dominated by plasma flow (Ktrans=Fp(1–Hct) where p is the tissue density and 1–Hct (haematocrit) is the plasma fraction). There are also regions where permeability is low compared with flow, although this is less common in extracranial tumours, and in these "permeability limited" circumstances, Ktrans=PS. The mixed situation occurs most commonly so neither flow nor permeability predominates; for extracellular gadolinium containing chelates there is a tendency for the influence of flow to outweigh that of PS in tumours. Evidence that Ktrans is dominated by flow in extracranial tumours is now emerging (see also limitations below). Recently, Kiessling et al reported a strong positive correlation between microbubble-enhanced Doppler ultrasound and dynamic T1 weighted DCE-MRI [77]. Previously, it has been shown that there is a near linear correlation between microbubble velocity measured on Döppler ultrasound and red blood cell velocity [78]. Both Lankester et al and Ah-See et al have shown strong positive correlations between Ktrans and rBF derived for T1 and T2* weighted DCE-MRI in pelvic and breast cancer, respectively [79] (Lankester K, pers. comm.). Further corroboration comes from the work of Maxwell et al who compared T1 weighted DCE-MRI enhancement parameters with tumour blood flow measured by the uptake of radiolabelled iodoantipyrine (IAP) in the rat P22 carcinosarcomas [80]. They showed that the time-course of changes in Ktrans and AUC as measured by DCE-MRI, and tumour blood flow rate measured by IAP uptake after treatment with a vascular targeting compound (Combretastatin — CA4P) were highly correlated, although the changes in Ktrans and AUC were smaller than those in blood flow by IAP. The application of DCE-MRI for monitoring antivascular anticancer treatments is discussed in detail below.
|
Limitations
|
|---|
Quantitative parameters are more complicated to derive compared with those derived semiquantitatively which deters their use at the workbench. Difficulties arise from more complex data acquisition requirements and by the lack of commercially available software to analyse acquired data. The model chosen may not exactly fit the data obtained (Figure 6) and each model makes a number of assumptions that may not be valid for every tissue or tumour type [16, 76]. There are uncertainties with regard to how reliable kinetic parameter estimates are, when derived from the application of tracer kinetic models to T1 weighted DCE-MRI data [81–83]. This is due to assumptions implicit in kinetic models and from assumptions made for the measurement of tissue contrast agent concentration [84]. For example, the Tofts' model uses a standard description of the time-varying blood concentration of contrast agent [85], and assumes that the supply of contrast medium is not flow limited and that tissue blood volume contributes negligibly to signal intensity changes compared with that arising from contrast medium in the interstitial space. As already noted above, this is not universally true in all parts of extracranial tumours. Thus, it is difficult to be certain about how accurately model-based kinetic parameter estimates compare with the physiological parameter that they purport to measure. Buckley has suggested that the application of commonly accepted models and their respective model-based assumptions to DCE-MRI data leads to systematic overestimation of Ktrans in tumours [86]. Despite these complexities it is important to remember that quantitative kinetic parameters can provide insights into underlying tissue pathophysiological processes that semiquantitative descriptors cannot. If the time varying contrast agent concentration can be measured accurately and the type, volume and method of administration of contrast agent are consistent, then it is possible to directly compare pharmacokinetic parameters acquired serially in a given patient and in different patients imaged at the same or different scanning sites. Furthermore, it is possible to use quantitative DCE-MRI as a tool for clinical decision making (see clinical experience below).
Validation
Many studies have correlated tissue enhancement with immunohistochemical microvessel density (MVD) measurements in a variety of tumours. Some MRI studies have shown broad correlations between T1 kinetic parameter estimates and MVD [77, 87–93] whereas others have found no correlation [63, 94, 95]. Recently, VEGF, a potent vascular permeability and angiogenic factor has been implicated as an additional explanatory factor that determines MR signal enhancement. Knopp et al reported that tumour vascular permeability to contrast media estimated by the rate constant kep on MRI closely correlated with tissue VEGF expression in breast tumours [96], whereas Su et al and Ah See et al have not [63, 97]. The importance of the role of VEGF in determining MR enhancement is supported by the spatial association of hyperpermeable capillaries, detected by macromolecular contrast enhanced MRI, and VEGF expression on histological specimens [98]. Furthermore, the observation that T1 weighted DCE-MRI measurements can detect changes in flow and permeability after anti-VEGF antibody after the administration of inhibitors of VEGF signalling, in xenografts [99–102] and in humans [58, 103, 104] lends weight to the important role played by VEGF in determining MR enhancement. Other characteristics that have been correlated with enhancement patterns include the degree of stromal cellularity and fibrosis [105, 106] tissue oxygenation [95, 107] and tumour proliferation [89, 108].
Clinical experience
Analysis of enhancement seen on T1 weighted DCE-MRI is a valuable diagnostic tool in a number of clinical situations. Its role for detecting subclinical disease remains to be determined and will probably include screening of subjects at high genetic risk of breast cancer or those put at risk from thoracic irradiation for the treatment of lymphoma. Screening women at genetic risk is the subject of several ongoing clinical trials [109, 110] and final reports from these studies are emerging [111, 112]. The most established role is in lesion characterization where it has been used for distinguishing benign from malignant breast and musculoskeletal lesions [64–68, 113]. Dynamic T1 weighted MRI studies have also been found to be of value in staging gynaecological malignancies, bladder and prostate cancers [114–117]. Recently, enhancement parameters have been shown to predict prognosis in patients with cervix cancers; that is, tumours with a fast initial rate of enhancement or vascular permeability were more likely to have a poorer prognosis [118] despite having a higher radiotherapy response rate [119]. DCE-MRI studies have also been found to be of value in detecting tumour relapse in the presence of fibrosis within treated tissues of the breast and pelvis [120–127]. Many studies now suggest that time since surgery and prior radiotherapy needs to be taken into account when interpreting images for the presence of recurrent disease in order to avoid interpretative errors [123, 124].
DCE-MRI is also able to predict response to or monitor the effects of a variety of treatments. These include neoadjuvant chemotherapy in bladder and breast cancers and bone sarcomas [128–131]. In breast cancer for example, it has been repeatedly shown that there is a progressive decrease in enhancement which accompanies tumour response to chemotherapy and that an increase or no change in enhancement predicts non-responsiveness [131–133]. Other treatments that can be monitored include radiotherapy in rectal and cervix cancers [134–137] androgen deprivation in prostate cancer [138] and vascular embolisation of uterine fibroids [139–141]. A full discussion of the clinical roles for DCE-MRI in evaluating antiangiogenic/antivascular drug treatments can be found below. It is noteworthy that enhancement on DCE-MRI can be affected by most types of successful treatments. This reflects on the fact that tumour cell kill, no matter how achieved, ultimately results in vascular shutdown, probably because of the loss of proangiogenic cytokine support which results in apoptosis of proliferating endothelial cells. DCE-MRI is thus limited in its ability to mechanistically inform on non-vascular processes that underlie tumour response to treatment although such information can be inferred from observing the timing of vascular responses. For example, DCE-MRI changes following chemotherapy are not observed within 24 h [142], whereas such early changes are seen for vascular targeting drugs [143]. DCE-MRI changes after anti-VEGF treatment can be observed as early as 48 h in humans [103, 104]. DCE-MRI changes after chemotherapy become notable after the first cycle and after 2 cycles can be used to predict efficacy of treatment [133].
|
DCE-MRI for evaluating antivascular cancer treatments
|
|---|
From an imaging perspective, it is more useful to consider antivascular drugs as being either antiangiogenic or vascular targeting agents (see article by Tozer in this issue). Antiangiogenic drugs prevent the formation of neovasculature and vascular targeting drugs (antiendothelial) cause rapid vascular damage resulting in haemorrhagic necrosis of tumours. Antiangiogenic drugs induce long-term changes in the tumour vasculature (mostly maturation) and are designed for continuous treatment. In contradiction, the effect on vasculature of vascular targeting drugs is reversible often within hours of a single treatment depending on the dose and thus intermittent administration is required.
Drugs that target VEGF or its receptor are the most common type of antiangiogenic agents. Anti-VEGF agents have action on either VEGF directly (e.g. anti-VEGF antibody or trap) or its receptor(s) in the extracellular (e.g. anti-VEGF receptor antibody) or intracellular domains (small molecules that target VEGF receptor tyrosine kinase). Effective inhibition of VEGF action would be expected to alter haemodynamic parameters such that they are detectable by DCE-MRI. These effects are mediated by the following mechanisms: (1) a direct antipermeability effect [144] (note that VEGF was previously named vascular permeability factor, VPF); (2) a decrease in local tissue levels of vasodilators (nitric oxide and prostacyclins) with relative early vasoconstriction; (3) a direct antiangiogenic effect leading to the formation of fewer immature (and thus fewer very leaky) vessels; (4) withdrawal of VEGF (a survival factor for proliferation endothelial cells) would lead to apoptosis of proliferating endothelial cells. It is considered that that mechanisms 1 and 2 probably occur early and explain the rapid changes seen on DCE-MRI in xenograft work [99] whereas mechanisms 3 and 4 act over a longer period where the overall effect is a reduction in total and functioning numbers of microvessels [145]. These antivascular effects are accompanied by significant micro-haemodynamic changes which included higher red blood cell velocities and increased blood flow within unaffected vessels [144].
Vascular targeting drugs cause rapid vascular damage by selectively targeting proliferating endothelial cells. Many of these agents bind to intracellular tubulin (the cytoskeleton of endothelial cells) and cause endothelial cell rounding and the formation of cell membrane blebs. These ultrastructural changes result in a sudden, marked increase in microvascular permeability and interstitial pressure with secondary vascular collapse. The effects of vascular targeting agents are reversible often within hours of a single treatment. Any effect on mature vasculature tends to occur as a secondary effect; longitudinal studies using DCE-MRI of antivascular treatment have shown decreasing efficacy with sequential doses of drug, a feature that might be interpreted as a vascular pruning effect [143]. The longer term effects of vascular targeting agents include a reduction in perfused vasculature, reduced perfusion, blood volume and vascular tortuosity.
A number of clinical studies have reported on the usage of DCE-MRI in phase 1 toxicity finding trials of antivascular cancer drugs [58, 103, 104, 143, 146]. The aims of these studies were similar; to determine safety and tolerability of repeated dosing and pharmacokinetic profile as well as to explore markers of biological activity, which includes DCE-MRI. Results from phase 1 clinical studies may also indicate which tumour types are best for future studies. For example the phase 1 work of Morgan et al [103] and efficacy studies of Yang et al [147] suggest that colorectal and renal cancers are particularly suited for anti-VEGF treatments. Hurwitz et al have taken such early evidence of biological activity and tumour susceptibility into chemotherapy combination trials with efficacy endpoints [148].
The use of DCE-MRI in phase 1 trials is to provide biological evidence of drug action and to impart confidence and thus accelerate the development of a particular drug or approach (go-no-go decisions). Good examples of the use of DCE-MRI in this way can be found in studies performed by Jayson et al, Morgan et al and Galbraith et al [103, 104, 149]. As a note of caution, it is important to remember that antivascular cancer therapies are designed to affect the development of small tumours during their growth phase when they have a higher proportion of immature, proliferating vessels. However, small tumours are not the traditional tumour type used in phase 1/2 clinical trials (usually larger, multiply treated and often multidrug resistant tumours with mature vasculature). An apparently efficacious drug may thus be considered ineffective because the wrong type of tumour was examined by DCE-MRI. Thus, the interpretation of angiogenesis imaging from phase 1 studies of antivascular drugs should be interpreted in the light of other evidence indicating vascularity directed action as elegantly demonstrated by Willett et al [150].
Before DCE-MRI is used in phase 1 studies, it is important to determine the correct timing for DCE-MRI evaluations in relation to the drug dosing schedule. Vascular targeting drugs act quickly (within hours, and have a reversible effects after a single dose) whereas anti-VEGF drugs are expected to act over the longer term. The timing of clinical studies can be guided by experimental xenograft data. Such information may be forthcoming from window chamber observations [144, 151] or from DCE-MRI studies [80, 99–102, 149, 152]. The results of xenograft studies can directly guide the timings of DCE-MRI in phase 1 clinical studies as demonstrated by Galbraith et al who performed DCE-MRI 4 h after the first dose to evaluate the vascular targeting agent Combretastatin following pre-clinical MRI experiments [149]. Although anti-VEGF drugs are expected to act over a longer period, pre-clinical data on anti-VEGF antibody showed that changes in microvessel permeability can occur as early as 90 min after the first dose [99, 100] and such data were able to inform on the usage of DCE-MRI at 2–4 days as well as at 4 week time point in human studies [104, 143, 153]. These points underscore the need to have good pre-clinical evaluation performed with a sufficient lead-time prior to clinical testing.
As already noted, the toxicity profile of vascular targeting and antiangiogenic drugs are quite different, and toxicity-based selection of dose may not be appropriate particularly for anti-VEGF drugs. DCE-MRI data may enable dosage and scheduling of treatment to be established for combination regimens. The "biologically active" dose can be indicated by imaging studies that demonstrate quantitative biological effects (e.g. by a reduction of blood flow or microvessel permeability) [103, 149]. Both Galbraith et al and Morgan et al showed that that there was a strong inverse correlation between a decrease in transfer constant and plasma levels of CA4P and PTK/ZK [103, 149] and such information leads to the selection of the dose of drug to be used in combination with other therapies for trials with efficacy endpoints. Morgan et al have also shown that it may be possible to subsequently predict antiangiogenic drug efficacy in patients with colorectal cancer on the basis of the degree of change in DCE-MRI [103].
|
MRI with macromolecular weight contrast media (MMCM)
|
|---|
Contrast agent kinetics
Low molecular weight contrast agents have a high first pass extraction fraction in both normal and tumour tissues [15]. Physiologists have noted that microvessels of cancers are specifically hyperpermeable to macromolecules [154, 155], whereas in normal tissues the endothelium prevents extravasation of large molecules such as plasma proteins. MMCM have molecular weights that can be as high as some serum proteins and are designed to have minimal first pass extraction fraction in normal vessels and therefore appear to be well suited for the measurement of tumour macromolecular hyperpermeability [15, 156, 157]. The ability of MMCM to characterize neovasculature is dependent on their molecular weight (MW), size and physiochemical properties. MMCM vary in MW between 5 kDa and 90 kDa or greater and both gadolinium and iron oxide containing agents are available. Following the intravenous administration of MMCM, 2 phases of enhancement can be observed on T1 weighted MRI during the first few minutes: bolus and steady-state phases [158]. During the bolus phase, signal enhancement arises from the contrast medium in the vascular compartment and reflects blood flow and volume. The steady state phase is usually observed after about a minute or two and is characterized by slow diffusion of contrast medium into the interstitial space.
In general, the rate of transfer of MMCM from the blood to the interstitial space (described by the rate constant KPS) is dependent on MW but this relationship is not linear. This has been shown by several workers including Lussanet et al who compared gadolinium based dendrimers with well defined MWs (between 0.7 kDa and 51 kDa) that were of equal chemical composition and molecular configuration [159]. The smallest agent (MW, 0.7 kDa) that they investigated was slightly larger than gadopentetate dimeglumine (Gd-DTPA) (MW, 0.5 kDa) and the largest agent (MW, 51 kDa; diameter, 5–6 nm) has a MW similar to serum albumin (MW, 70–90 kDa; diameter, 6 nm) [160]. They showed that there was an inverse relationship between MW and KPS such that doubling MW reduces KPS by 25% (de Lussanet QG, pers. comm.). Plasma clearance of MMCM is also dependent on MW with the largest agents being retained longest. Safety is of major concern in designing MR contrast agents. One of the reasons for the success of ECF contrast agents, aside from their paramagnetic properties, is their minimal interaction with tissues, stability and fast clearance. Full clearance of the paramagnetic gadolinium ions is crucial for reasons of toxicity. Designing complex high MW contrast agents must also meet these safety criteria.
Albumin-(Gd-DTPA)30 is the prototype MMCM but this agent has been found to be immunogenic and there is significant retention of gadolinium in the liver and bone [161] which limits its clinical use. Polylysine-(Gd-DTPA) is not readily biodegradable, which also makes it unsuitable for human use. Recently USPIO particles (diameter 10–30 nm) have been investigated as MMCM [93, 102, 162, 163]. USPIO MMCM persist in the circulation with half-lives of between 2 h and 24 h depending on the preparation [164]. Human studies with USPIOs as macromolecular contrast media capable of characterizing tumour permeability have begun to be reported [165]. Signal to noise ratio and degree of enhancement are greater with increasing doses of USPIO contrast medium [162]. However when injected as a bolus at higher doses, USPIO MMCM may not be tolerated well. Furthermore, at higher doses, T2* effects can overwhelm the desired T1 enhancement effects in tissues. T2* effects can be marked when vascular densities are high and heterogeneously distributed. However, both albumin-(Gd-DTPA)30 and USPIO enhanced MRI suffer from very low absolute signal enhancement due to low rates of contrast medium extravasation. Consequently, long scans periods are needed to observe enhancement above the image noise level. In contradistinction, intermediate-sized MMCM agents such as P792 (MW, 7 kDa) and Gadomer-17 (MW, 17 kDa) have good safety profiles, yield images of good signal-to-noise ratio and have already proceeded to advanced clinical trials as agents for MR angiography (NOT angiogenesis imaging) [166, 167]. The utility of intermediate molecular weight agents is considered in more detail below.
Data acquisition and quantification
A timed series of T1 weighted images is used to detect and measure MMCM hyperpermeability. Ideally, both the tumour and the blood pool are monitored simultaneously [157, 168]. All studies assume that the difference between pre-contrast and post-contrast T1 relaxavity values at each time point (
R1) is assumed to be proportionate to the concentration of contrast medium both in vessels and in tissues [169]. Numerous approaches have been proposed for analysis of the MRI data. Some approaches ignore the input function while others include the blood response curve to correct for variations in injection speed, contrast elimination and circulation time. The application of a two compartment, bidirectional or unidirectional kinetic model to the dynamic response data enables the estimation of the transfer coefficient of endothelial permeability (KPS, units ml min–1 per volume of tissue) and the (flowing) fractional plasma volume (fPV, ml per volume of tissue) [168, 170].
KPS, which represents the rate of transfer of contrast medium from the blood to the interstitial space, is highly dependent on the permeability and surface area of the endothelium. Unlike low MW contrast agents (which have a large first pass extraction – hence KPS values are dominated by flow in tumours), high MW contrast agents leak from the blood more slowly, and their transfer into the interstitial matrix is less dependent on tumour blood flow. Thus, for MMCM, KPS values will approximate PS product. KPS of intermediate molecular weights is influenced by both permeability surface area product and plasma flow. fPV can be measured with a high MW MR contrast agent because of relatively long intravascular retention of the contrast medium (although poor signal to noise ratio can be problematic for some agents). USPIOs contrast media tend to overestimate fPV because magnetic field inhomogeneities (from their large paramagnetic moments) extend beyond the vascular compartment. For intermediate size MMCM, there is a tendency towards overestimated fPV because of diffusion out of the vascular compartment (loss of compartmentalization) during the time period of the experiment; this is even truer for ECF contrast agents.
Validation and suitability for monitoring antivascular treatments
Xenograft experiments have validated MMCM enhanced MRI for characterizing tumour angiogenesis. An early study using albumin-(Gd-DTPA)30 showed significant hyperpermeability of tumours compared with normal tissues in a rat fibrosarcoma model [171]. Using quantification techniques, van Dijke et al showed a strong positive correlation between both tumour KPS and fPV with histological microvessel density (MVD) in two subtypes of mammary R3230 adenocarcinoma, one slow growing and the other fast and aggressive [172]. In contradistinction, Helbich et al who studied the same tumour model (mammary R3230 adenocarcinoma) with DCE-MRI with an ECF contrast agent and showed no significant correlations between enhanced parameters (except for T2* weighted DCE-MRI) and histological MVD [173]. USPIO-enhanced MRI studies have also reported good correlation between KPS and histological tumour grade [163] and MVD [93]. These correlations were essentially the same as those obtained using albumin-(Gd-DTPA)30 in the same tumour [162, 163].
Several xenograft studies have now shown that intermediate MW contrast media in the range of 10–30 kDa have poor correlation with tumour histological grades or microvessel density [174–178]. For example, Su et al evaluated three contrast agents of differing MW (Gd-DTPA (<1 kDa), Gadomer-17 (17 kDa) and albumin-(Gd-DTPA)30 (70–90 kDa)) to study breast tumours in xenografts [177]. They confirmed that Gd-DTPA was able to distinguish benign from malignant tumours but was unable to grade malignant tumours. They found that Gadomer-17 was able to distinguish between high and low grade malignant tumours but was unable to distinguish low grade cancers from benign lesions. Albumin-(Gd-DTPA)30 was able to distinguish three but low contrast-to-noise ratio in the images was a major technical concern.
Once approved for human use, MMCM assays of tumour microvascular characteristics may be able to address a number of important clinical issues including non-invasive characterization of tumours. Strong pre-clinical evidence is beginning to emerge that MMCM enhanced MRI can monitor antivascular cancer treatments which should translate relatively easily into human studies. Pham et al demonstrated that when monoclonal human anti-VEGF antibody was administered to athymic rats implanted with human MB-MDA-435 breast adenocarcinoma, large reductions in KPS (measured by albumin-(Gd-DTPA)30 enhanced MRI) were seen within 24 h after treatment. Reductions in KPS occurred without a change in fPV [99]. However, as already noted albumin-(Gd-DTPA)30 is not suitable for human use. Recently, Turetschek et al reported on the usefulness of MMCM enhanced MRI in MD-MBA-435 xenografts treated with PTK787/ZK, a VEGF receptor tyrosine kinase inhibitor [102]. They compared albumin-(Gd-DTPA)30 (6.0 nm diameter) with an USPIO, 30 nm in diameter with MRI scans done 7 days apart. Although significant growth delays were seen in treated animals, KPS values declined only slightly (significant increases in KPS was seen in control animals). However, a detailed look at the results reveals that 8 of 10 treated animals and 1 of 10 control animals had reductions in KPS values. Estimated values for fPV did not differ significantly between treatment groups despite a reduction in microvessel density counts. The lack of a significant reduction of KPS values in treated animals is a cause for concern when translating these results into human studies where a significant reduction of KPS would be required to gauge efficacy of drug action. As already noted, large molecular weight contrast media also have the additional disadvantages of low contrast-to-noise ratio in the images obtained and are long experiments to undertake in humans (taking 40–50 min on average in animals for the dynamic sequence).
As intermediate MW MMCM are likely to achieve clinical approval soon and yield images of good signal to noise ratio in a relatively short period of time (5–7 min); these agents have also been investigated for their ability to monitor antiangiogenesis treatments. Turetschek et al noted that medium-sized contrast media (such as ZK181220 (MW, 25.0 kDa) and Gadomer-17) tended to yield higher estimates of fPV and KPS compared with albumin-(Gd-DTPA)30 and USPIOs but could still allow MR-based monitoring of anti-VEGF antibody-mediated drug effects (on the basis of reductions in fPV and KPS) in human MB-MDA-435 breast adenocarcinomas xenografts treated for 7 days [179]. These effects were observed despite the fact that there was a tendency to overestimate fPV (compared with albumin-(Gd-DTPA)30) which as noted above is due to partial extravasation [159]. Similar results were found by Pradel et al and Bradley et al who evaluated the contrast agent P792 in xenografts treated with the small molecule tyrosine kinase receptor inhibitors acting against the VEGF receptor. Both groups showed significant reductions in KPS and fPV within 1–3 days of commencing treatment [180, 181]. These data taken together seem to indicate that dynamic MRI enhanced with medium-sized MR contrast media should be able to monitor effects of anti-VEGF therapies clinically on account of their kinetic characteristics, good safety profile, good signal to noise ratio of images, short imaging times and suitability for quantification.
As intermediate and high MW contrast media will become available soon, so it has become necessary to comment on their relative potential utility compared with low MW contrast agents. As noted above, the flow component intrinsic in the measured KPS is dependent on the MW of the contrast agent. Relative reductions in KPS resulting from increased MW of the contrast agent are in effect reductions of the flow contribution. The relationship between KPS and MW predicts that the largest gain in overcoming flow contamination in KPS is obtained by moving from the range of low to intermediate MW agents, and less so by moving from intermediate to high MW agents. It is most likely that KPS, rather than the fPV will be of main interest for indexing angiogenic activity and effect of antivascular treatment. Thus, ECF contrast media will continue to play an important role when the treatment effects involve reducing the magnitude and or heterogeneity in tumour blood flow, since KPS becomes highly sensitive for such changes in blood flow when using low MW contrast agents.
|
Intrinsic susceptibility contrast or BOLD MRI
|
|---|
Intrinsic susceptibility contrast and BOLD MRI are used interchangeably to acknowledge the primary source of contrast in BOLD MR images, endogenous deoxyhaemoglobin. Deoxyhaemoglobin, which is paramagnetic, creates susceptibility variations in the magnetic field, increasing the MR transverse relaxation rate R2* (=1/T2*) of water in blood and in the tissue surrounding blood vessels. Deoxyhaemoglobin therefore acts as an intrinsic contrast agent [11]. Gradient recalled echo (GRE) sequences are sensitive to R2* and are thus sensitive to blood deoxyhaemoglobin levels which itself is a function of the vascular volume of tumour perfused by red blood cells. The contrast in GRE images acquired at a particular echo time (TE) is dependent on tissue flow and oxygenation. Multigradient echo techniques allow the calculation of relaxation (R2*) maps that have a greater sensitivity to tissue oxygenation. The primary advantage of BOLD MRI techniques is that there is no need to administer contrast material. Measurements can be repeated as needed with almost no limitation. BOLD contrast is not sensitive to fluctuations in vascular permeability. A major reservation for intrinsic contrast imaging is the low contrast to noise ratio in the images obtained. Taylor et al have also reported that human BOLD studies with carbogen (5% CO2: 95% O2) vasomodulation are technically challenging with a high failure rate [182].
Vascular function can be evaluated by observing changes in BOLD signal intensity in response to hyperoxia and hypercapnia [11, 12, 183]. Clinical applications of BOLD MRI with carbogen inhalation has revealed high signal enhancements in a variety of tumours including head and neck carcinomas [182]. Signal enhancements observed in response to carbogen occur due to changes in blood flow and oxygenation. Abramovitch et al have shown that vascular responsiveness to hyperoxia and hypercapnia is also dependent on the maturity of the vasculature of the tumour being studied [12]. Mature blood vessels have pericyte and smooth muscle covering and are thus able to respond to vasoactive stimuli in a way that immature vessels cannot. The differential response to hyperoxia and hypercapnia can thus be used to map blood vessel maturation and function and their differential sensitivity to changes in VEGF expression [12]. BOLD imaging can also be used to monitor the development and regression of tumour angiogenesis [184, 185]. These results suggest that BOLD MRI may be able to monitor the effects of antiangiogenesis drugs although no report of this usage has appeared in the human literature to date.
Gross et al have recently reported that BOLD MRI can detect changes in oxygen consumption and subsequent vascular shutdown induced by light stimulation of a novel photodynamic therapy agent (TOOKAD) which acts as a vascular targeting therapy [13]. This effect was seen within a few minutes after light exposure. They noted an increase in R2* when either blood doped with TOOKAD or when melanoma xenografts treated with TOOKAD were exposed to light. The BOLD effect (25–40% reduction of MR signal intensity) was shown to be ascribable to intravascular photosensitization of TOOKAD leading to photochemical oxygen consumption (75% reduction of serum PO2) and by local occlusion of vessels and stasis of blood flow demonstrated on intravital microscopy.
Robinson et al have investigated the use of BOLD MRI to assess the efficacy of a vascular targeting agent ZD6126 in rodent tumours. ZD6126 is a tubulin binding agent with a mechanism similar to Combretastatin-A4-phosphate causing the selective destruction of proliferating tumour blood vessels, cessation of tumour blood flow, death of tumour cells due to nutrient starvation, and massive tumour necrosis [186]. The hypothesis tested by Robinson et al was that following treatment with ZD6126, haemoglobin within erythrocytes would deoxygenate, resulting in an increase in tumour R2* as suggested by the results of Gross et al [13]. Robinson et al showed that tumour R2* significantly increased for the first half hour after challenge with ZD6126 and this coincided with a decrease in tumour perfusion as indicated by uptake of Hoechst 33342 [187]. Early human tumour work shows similar effects (Figure 7). Interestingly, tumour R2* significantly decreased 24 h post-treatment with ZD6126, which also correlated with a significant decrease in tumour perfusion measured by Hoechst 33342 uptake [14]. The decrease in tumour R2* observed at 24 h could be due to one of several factors including the development of oedema and/or a complete lack of delivery of red cells as part of massive central necrosis assessed histologically. Further work is required to elucidate the mechanisms responsible for the changes observed; however it does seem likely that changes in tumour R2* may prove to be a simple and convenient biomarker for detecting acute changes induced by antivascular therapies.
View larger version (86K):
[in this window]
[in a new window]
|
Figure 7. Significant changes in blood oxygenation level dependent (BOLD) signal intensity following antivascular targeting treatment. Axial BOLD spoiled gradient echo MR images (TR=100 ms, TE=60 ms, flip angle=40°) through the mid-pelvis shows a large, left inferior hypogastric lymph node (arrow) in a patient with malignant peritoneal carcinoma. Images were obtained twice before (–5 days and –1 day) and 4 h and 24 h after the first administration of a vascular targeting agent (Combretastatin-A4-phosphate, 52 mg m–2). Darkening within the centre of the lesion at 4 h with treatment is consistent with the presence of reduced blood flow and an increase in the amount of deoxyhaemoglobin. This BOLD effect reverses by 24 h. Corresponding signal intensity time curves are shown which show marked but reversible alteration in contrast agent kinetics.
|
|
|
Challenges for MRI in antivascular treatment evaluations
|
|---|
The great variety of MRI techniques that are able to acquire information on tumour microvasculature has resulted in a number of approaches being used in the clinic. This variety can make it difficult to make meaningful comparisons between different techniques, tissue types and to compare data obtained from different imaging centres. If MRI is to be used for the selection of antivascular drugs that advance into efficacy trials then it will be necessary to develop standardized approaches to measurement, and robust analysis approaches with clear accepted endpoints specified prospectively that have biological validity.
For DCE-MRI with low molecular weight contrast medium enhancement it is recognised that high resolution and short imaging times are competing examination strategies on current equipment and software. Higher temporal resolution imaging necessitates reduced spatial resolution, decreased anatomical coverage or both. Higher temporal resolution techniques are essential for DCE-MRI techniques and appear to improve specificity of T1 weighted examinations because of better characterization of tissue enhancement. For MMCM enhanced MRI, high temporal resolution is not as stringent a requirement. Even though data collection procedures for quantitative examinations differ from those used in routine clinical practice, there is debate as to which technique(s) is/are best [188–190]. The MRI community needs to agree on a limited number of examination and analysis protocols in order to enable techniques to be validated and used in clinical trials.
Another issue that needs to be addressed is that of data collection in body parts where there is a large degree of physiological movement such as the lungs and liver. The presence of motion can invalidate functional vascular parameter estimates particularly for pixel-by-pixel analyses. Methods for overcoming/minimizing these effects include the application of navigator techniques [191] or imaging in the non-axial plane using sequential breat
Top
Abstract
Introduction
) into contrast agent concentration (•) is performed first according to the method described by Parker et al [