**4.2.2 Dynamic Contrast Enhancement**

Dynamic contrast-enhanced T1W MRI (DCE-MRI) was developed in the 1990s for estimating BBB leakage (Tofts and Kermode 1991). It is entirely an approach to measure perfusion based on the T1 shortening caused by the infusion of a paramagnetic contrast media. The main objective of this technique is to measure tumor permeability by dynamic T1W imaging after the intravenous administration of a contrast medium. Typically, a single lower dose (0.1 mmol/Kg) and a lower rate than in DSC studies is injected to the patient followed by repetitive image acquisitions (every 15-30 seconds) through the tumor up to 5- 10 minutes rather than the first pass of the bolus. The method is focussed in allowing the leakage of the CA into the extravascular space and coming into equilibrium after several passes of the bolus through the microvascular tumor bed.

In this sense, one of the main advantages of this method is that the extravasation of the contrast material through the BBB is not an artefact to be corrected. Dynamic T1W MRI acquisitions afford better visualization of tumoral heterogeneity in gliomas, improving the accuracy of the technique to detect tumor staging, tumor recurrence and follow-up response to therapies.

Appropriate information from DCE images requires measuring the real T1 values of the tissues to take them as a baseline to perform an accurate pharmacokinetic analysis (Evelhoch 1999). To get an optimal temporal resolution, three dimensional imaging schemes are typically used obtaining the arterial input function in the centre of the acquisition volumes. Hence, data acquisition parameters may influence data analysis and it is necessary to optimize them. Several post-processing methods have been developed to measure perfusion in gliomas using DCE approaches. M. Pauliah et al. described an improved procedure to perform cerebral perfusion maps using a spoiled gradient-recalled echo sequence and reporting not only morphological but also functional information in a single imaging acquisition (Pauliah, Saxena et al. 2007). Authors estimate a quantitative evaluation of both regional CBV and CBF using singular value decomposition (SVD)-based deconvolution techniques in which the concentration of contrast material passing through a given voxel can be expressed as a convolution of the AIF with the residue function tissue concentration (Ostergaard, Sorensen et al. 1996; Ostergaard, Weisskoff et al. 1996). Figure 10 shows one particular case from this study.

In summary, DCE is a powerful tool capable of providing quantitative assessment of contrast uptake and characterization of microvascular structure in human gliomas. Even more, several comparative studies have reported that DCE MRI studies of tumour enhancement in glioma patients can be used for repeated studies with reasonable reproducibility (Jackson, Jayson et al. 2003).

of interest. This is one factor that limits the ability to quantitate hemodynamic parameters from DSC studies. To measure AIF the imaging planes through the brain have to be chosen to contain a major intracranial artery feeding the tissue, typically one paraclinoid internal carotid artery or the middle cerebral artery (Covarrubias, Rosen et al. 2004). Then, a deconvolution algorithm is used to post-process the data and generate CBF and MTT maps. However, the absolute determination of hemodynamic parameters in not a crucial factor in grading human gliomas, the resulting ratios being equally valid for this purposes since they

Dynamic contrast-enhanced T1W MRI (DCE-MRI) was developed in the 1990s for estimating BBB leakage (Tofts and Kermode 1991). It is entirely an approach to measure perfusion based on the T1 shortening caused by the infusion of a paramagnetic contrast media. The main objective of this technique is to measure tumor permeability by dynamic T1W imaging after the intravenous administration of a contrast medium. Typically, a single lower dose (0.1 mmol/Kg) and a lower rate than in DSC studies is injected to the patient followed by repetitive image acquisitions (every 15-30 seconds) through the tumor up to 5- 10 minutes rather than the first pass of the bolus. The method is focussed in allowing the leakage of the CA into the extravascular space and coming into equilibrium after several

In this sense, one of the main advantages of this method is that the extravasation of the contrast material through the BBB is not an artefact to be corrected. Dynamic T1W MRI acquisitions afford better visualization of tumoral heterogeneity in gliomas, improving the accuracy of the technique to detect tumor staging, tumor recurrence and follow-up response

Appropriate information from DCE images requires measuring the real T1 values of the tissues to take them as a baseline to perform an accurate pharmacokinetic analysis (Evelhoch 1999). To get an optimal temporal resolution, three dimensional imaging schemes are typically used obtaining the arterial input function in the centre of the acquisition volumes. Hence, data acquisition parameters may influence data analysis and it is necessary to optimize them. Several post-processing methods have been developed to measure perfusion in gliomas using DCE approaches. M. Pauliah et al. described an improved procedure to perform cerebral perfusion maps using a spoiled gradient-recalled echo sequence and reporting not only morphological but also functional information in a single imaging acquisition (Pauliah, Saxena et al. 2007). Authors estimate a quantitative evaluation of both regional CBV and CBF using singular value decomposition (SVD)-based deconvolution techniques in which the concentration of contrast material passing through a given voxel can be expressed as a convolution of the AIF with the residue function tissue concentration (Ostergaard, Sorensen et al. 1996; Ostergaard, Weisskoff et al. 1996). Figure 10 shows one

In summary, DCE is a powerful tool capable of providing quantitative assessment of contrast uptake and characterization of microvascular structure in human gliomas. Even more, several comparative studies have reported that DCE MRI studies of tumour enhancement in glioma patients can be used for repeated studies with reasonable

incorporate the measurement errors from both values.

passes of the bolus through the microvascular tumor bed.

**4.2.2 Dynamic Contrast Enhancement** 

to therapies.

particular case from this study.

reproducibility (Jackson, Jayson et al. 2003).

Fig. 10. DCE images of a patient with histologycally proved astrocytoma grade III. A: Post Gd-injection T1W image. B: T2W image. D: T1W image (without CA administration). C: CBV and E: CBF maps generated

#### **4.2.3 Arterial Spin Labelling**

ASL perfusion imaging in the brain adds a new functional dimension to non invasive evaluation of the brain functionality with MRI. It is the only method that yields quantitative monitoring of cerebral perfusion over a period of several hours. Its non invasive nature allows an unlimited number of repeated measurements to be made with a high temporal resolution. ASL uses spatially selective inversion of the inflowing arterial blood magnetization as a method to label water molecules in blood. So, in a typical experiment the arterial blood is tagged and after a given delay, this tagged blood arrives at the imaging plane and an image is acquired. The MRI signal from inverted magnetization is made negative relative to uninverted one. When the labelled blood reaches the tissue, it attenuates the pre-existing signal from the image of that tissue. Therefore, a control measurement is required without tagging the arterial blood. The difference in signal intensity between both images is proportional to rCBF, through a calibration factor that depends primarily on global properties of blood. A schematic diagram of the process is shown in figure 11. This difference signal measures how much of the original arterial magnetization created by the inversion pulse has been delivered to the voxel and survives to the time of measurement. The magnetization difference (control minus taged) carried out into the voxel can be conceptually similar to a contrast material delivered to it by CBF. Under this assumption the same kinetic modelling arguments used in bolus tracking or DSC studies can be applied to ASL (Buxton, Frank et al. 1998). Nevertheless, several premises remain to be taken into account for an accurate

ASL has been also used as a functional method to detect CBF changes in local CBF during task activation as a surrogate marker or neuronal activity. This can be very helpful in the assessment of gliomas. To investigate if ASL methodologies contribute measurements that can be used as reliable markers of tumor grade, several studies have compared the perfusion results obtained by MRI with histological data in glioma patients (Warmuth, Gunther et al. 2003; Wolf, Wang et al. 2005). As perfusion MR leads to images of the entire neoplasm, these can help in characterizing the malignancy in gliomas overcoming some of the limitations of histopathology sampling. In this line, ASL perfusion images allow delineating tumor heterogeneity and this can be used to guide biopsy towards the most malignant region of the tumor. Some examples of perfusion maps obtained from ASL acquisitions in gliomas are

Fig. 12. Differential diagnosis of gliomas using ASL perfusion. A: Giant cells glioblastoma. B: Anaplastic astrocytomas. In both, left: post-contrast T1W; right: perfusion map from PASL

A close relationship has been established between ASL and bolus tracking methods in determining CBF in brain tumors, been both approaches able to differentiate between high-grade and low-grade gliomas. Probably, the main limitation of using ASL sequences is that they only allow for the calculation of CBF, whereas MR perfusion measurements based on the administration of a contrast material can provide additionally, CBV and MTT values by using DSC methods (T2\*W based images), and permeability values by DCE sequences (T1W based images). This additional information may contribute appreciably to the characterization and grading of gliomas. On the other hand, comparative studies in glioma patients have shown that the ASL technique is more suitable than exogenous contrast-based perfusion analysis for assessing hypervascularized lesions, whereas it has a bigger error rate in the assessment of

Current methods to grade gliomas have inherent limitations, the infiltrative proliferation as an intrinsic feature of the tumor becoming a significant problem. Conventional MR images with CA provide important information regarding Gd-based enhancement that is useful in characterizing tumor grade. The most malignant enhancing brain tumors include

(Tourdias, Rodrigo et al. 2008). Reproduced with permission.

hypovascularized ones (Warmuth, Gunther et al. 2003).

**4.3 Tumor grading by perfusion MRI** 

shown in Figure 12.

quantification of CBF by using ASL. The method necessarily has to: i) perform an accurate control measurement; ii ) create a well-defined tagged bolus of arterial blood, waiting sufficiently long for the bolus to be delivered to the imaging plane; and iii) account for relaxation of the magnetization label (Buxton 2005).

Fig. 11. Fundamentals of the ASL method: Arterial blood is tagged by inverting the magnetization and, after a delay, it arrives to imaging plane decreasing its intensity in manner proportional to blood flow (taken from "ASL Perfusion Imaging: Concepts and Applications", David C. Alsop., ISMRM-2006). Reproduced with permission.

ASL is an attractive and ingenious method for clinical applications because it does not use an intravenous contrast agent, but merely a manipulation of the proton spin of water molecules within the vasculature, that has the potential to provide a robust estimation of CBF. Hence, the agent used in ASL for assessing perfusion is labelled water and not a paramagnetic or superparamagnetic contrast medium, potentially affecting the physical, chemical or physiological properties of the blood. Another advantage of this approach is that it is independent of tumor permeability, without the need to perform additional corrections in the post-processing of the data. The perfusion map obtained by difference of images is available just after acquisition with minimal post-processing, representing a clear benefit in routine clinical settings. On the other hand, ASL leads reliable absolute quantification of CBF that is not affected by BBB damages.

The ASL methods apply a powerful magnetic gradient to invert the inflowing blood magnetization tagging it as it flows upstream. ASL techniques can be divided in two main classes, depending on how the tag is applied: continuous (CASL) or pulsed (PASL). In CASL the inflowing arterial blood is continuously labelled bellow the slab to be imaged until the tissue magnetization gets a steady state (Petersen, Zimine et al. 2006). In PASL the spin inversion is achieved using short inversion pulses in a thick slab located next to the slice of interest. This method is easier to implement and has less practical problems as compared with CASL (Petersen, Zimine et al. 2006). Interestingly, PASL does not suffer as many magnetization transfer effects as CASL methods, and the selective inversion applied is closer to the image slices yielding smaller signal attenuation because of the reduced transit delay and a better inversion efficiency. Even more, short inversion pulses minimize problems of energy deposition in humans at high fields.

quantification of CBF by using ASL. The method necessarily has to: i) perform an accurate control measurement; ii ) create a well-defined tagged bolus of arterial blood, waiting sufficiently long for the bolus to be delivered to the imaging plane; and iii) account for

Fig. 11. Fundamentals of the ASL method: Arterial blood is tagged by inverting the magnetization and, after a delay, it arrives to imaging plane decreasing its intensity in manner proportional to blood flow (taken from "ASL Perfusion Imaging: Concepts and

ASL is an attractive and ingenious method for clinical applications because it does not use an intravenous contrast agent, but merely a manipulation of the proton spin of water molecules within the vasculature, that has the potential to provide a robust estimation of CBF. Hence, the agent used in ASL for assessing perfusion is labelled water and not a paramagnetic or superparamagnetic contrast medium, potentially affecting the physical, chemical or physiological properties of the blood. Another advantage of this approach is that it is independent of tumor permeability, without the need to perform additional corrections in the post-processing of the data. The perfusion map obtained by difference of images is available just after acquisition with minimal post-processing, representing a clear benefit in routine clinical settings. On the other hand, ASL leads reliable absolute

The ASL methods apply a powerful magnetic gradient to invert the inflowing blood magnetization tagging it as it flows upstream. ASL techniques can be divided in two main classes, depending on how the tag is applied: continuous (CASL) or pulsed (PASL). In CASL the inflowing arterial blood is continuously labelled bellow the slab to be imaged until the tissue magnetization gets a steady state (Petersen, Zimine et al. 2006). In PASL the spin inversion is achieved using short inversion pulses in a thick slab located next to the slice of interest. This method is easier to implement and has less practical problems as compared with CASL (Petersen, Zimine et al. 2006). Interestingly, PASL does not suffer as many magnetization transfer effects as CASL methods, and the selective inversion applied is closer to the image slices yielding smaller signal attenuation because of the reduced transit delay and a better inversion efficiency. Even more, short inversion pulses minimize problems of

Applications", David C. Alsop., ISMRM-2006). Reproduced with permission.

quantification of CBF that is not affected by BBB damages.

energy deposition in humans at high fields.

relaxation of the magnetization label (Buxton 2005).

ASL has been also used as a functional method to detect CBF changes in local CBF during task activation as a surrogate marker or neuronal activity. This can be very helpful in the assessment of gliomas. To investigate if ASL methodologies contribute measurements that can be used as reliable markers of tumor grade, several studies have compared the perfusion results obtained by MRI with histological data in glioma patients (Warmuth, Gunther et al. 2003; Wolf, Wang et al. 2005). As perfusion MR leads to images of the entire neoplasm, these can help in characterizing the malignancy in gliomas overcoming some of the limitations of histopathology sampling. In this line, ASL perfusion images allow delineating tumor heterogeneity and this can be used to guide biopsy towards the most malignant region of the tumor. Some examples of perfusion maps obtained from ASL acquisitions in gliomas are shown in Figure 12.

Fig. 12. Differential diagnosis of gliomas using ASL perfusion. A: Giant cells glioblastoma. B: Anaplastic astrocytomas. In both, left: post-contrast T1W; right: perfusion map from PASL (Tourdias, Rodrigo et al. 2008). Reproduced with permission.

A close relationship has been established between ASL and bolus tracking methods in determining CBF in brain tumors, been both approaches able to differentiate between high-grade and low-grade gliomas. Probably, the main limitation of using ASL sequences is that they only allow for the calculation of CBF, whereas MR perfusion measurements based on the administration of a contrast material can provide additionally, CBV and MTT values by using DSC methods (T2\*W based images), and permeability values by DCE sequences (T1W based images). This additional information may contribute appreciably to the characterization and grading of gliomas. On the other hand, comparative studies in glioma patients have shown that the ASL technique is more suitable than exogenous contrast-based perfusion analysis for assessing hypervascularized lesions, whereas it has a bigger error rate in the assessment of hypovascularized ones (Warmuth, Gunther et al. 2003).

### **4.3 Tumor grading by perfusion MRI**

Current methods to grade gliomas have inherent limitations, the infiltrative proliferation as an intrinsic feature of the tumor becoming a significant problem. Conventional MR images with CA provide important information regarding Gd-based enhancement that is useful in characterizing tumor grade. The most malignant enhancing brain tumors include

permeability in gliomas can be used as a non invasive surrogate marker able to provide information related to rate of angiogenesis in the tumor. High-grade gliomas depict high vascular permeability, a consequence of the massive BBB disruption leading to extravasation of contrast material into the extravascular space that becomes detectable in

Although studies in humans have been typically carried out with DCE methods, it is also possible to perform and compare permeability measurements in grading gliomas employing T2\* DSC techniques, within certain limitations (Provenzale, Wang et al. 2002). In this case, the Ktrans maps can be generated by employing a calculation that allows the separation of T1 effects from the analysis. The advantage of using this method is that it is possible to obtain CBV and permeability maps from the same data set, providing simultaneously, two types of

At present, the predominant method to assess tumor response to treatment is to measure the size of the lesion by using conventional, paramagnetically enhanced MRI. However, tumor size is not always specific of tumor behaviour and neoplastic features associated with aggressiveness of the tumor may vary without changes in the tumor size. On the other hand, since angiogenesis plays a critical role not only in the development of brain tumors but also in the malignant transformation, many of the novel therapies currently investigated are focused on antiangiogenic strategies. Monitoring the success of such therapeutic methods relies on the ability to quantify tumor vascularization through the measurements of hemodynamic parameters. It is widely accepted now that the use of quantitative measurements of tissue enhancement characteristics provides a more valuable indicator to monitor the therapeutic response of tumours to novel antiangiogenic drugs in the treatment of gliomas. For example, rCBV is supposed to reflect microvessel density, an early response in this parameter to antiangiogenic therapies should show a decrease, thus providing the

Figure 13 shows a clear example of this behaviour. While T1W images show tumor growth even 6 weeks after initiating the treatment, rCBV maps from PWI depict a much earlier drop in tumor perfusion. This demonstrates that tumor size and perfusion may lead opposite

Because the increase in endothelial permeability is a primary effect of vascular endothelial growth factor (VEGF) and other angiogenic cytokines, changes in the capillary leakage of contrast media may offer a predictable response to anti-angiogenic therapies. Quantification of the enhancement effect is therefore a good candidate as a potential biomarker of drug activity. Changes in permeability occur rapidly in response to VEGF inhibitions, so assessment of these changes either by DSC or DCE can be used as a marker of antiangiogenic therapy responses. Although quantification of permeability would be an optimal strategy, this is in fact, extremely difficult to achieve. Along these lines, several pharmacokinetic models to analyze dynamic enhancement data have been developed to estimate Ktrans and used it as an therapeutic indicator (Jackson, Jayson et al. 2003). Now, it is accepted and recommended the use of perfusion MRI to monitor tumor angiogenesis and evaluate the response to antiangiogenic therapies, a subject that has gained considerable clinical and research efforts, even leading to published recommendations concerning this

**4.5 Assessment of treatment response with perfusion MRI** 

conclusions because they represent indeed, different tumor features.

potential to predict growth or regression.

important capacity of MRI (Brasch, Li et al. 2000).

MR images.

valuable information.

high-grade gliomas, metastases, and lymphomas. These methods assign the most aggressive and malignant grade to the most enhanced tumors, but regarding the gliomas, this is not always reliable. A high-grade glioma may be often mistaken for a low-grade one when images show minimal edema, no gadolinium enhancement, no necrosis and no mass effect. Equally, low-grade gliomas can depict sometimes peritumoral edema, contrast enhancement, necrosis and mass effect leading to a high-grade false diagnosis (Law, Yang et al. 2003). In histopathology grading of gliomas, tumor cellularity and vascularity are the major factors to be taken into account. This information is accessible by the advanced functional MRI methodologies detailed in this chapter and can help in brain tumor grading. As regarding perfusion-based maps it is widely accepted that high regional CBV is more likely in high-grade gliomas. This parameter can reflect tumor vascularity and be used as an indicator of histological grade. In fact, many groups have shown the ability of DSC MRI for estimating the histopathological glioma grade. Several studies reported rCBV values measured in a large number of brain tumor patients suggesting that this parameter can improve distinction between high-grade and lowgrade gliomas (Aronen, Gazit et al. 1994; Calli, Kitis et al. 2006) with a high true-positive rate and a very low false-negative rate. On the other hand, gliomas have a great histopathological variability and in this sense, perfusion maps allow also for the localization of foci with very high rCBV values within the tumor. PWI can thus be very helpful not only for a valuable preoperative grading, but also for guiding the surgeon to the most appropriate site to obtain a representative biopsy of the most malignant portion. Although CBV is the most commonly used hemodynamic parameter in glioma grading, the largest limitation, entailing special clinical relevance, is the lack of discrimination between anaplastic gliomas and glioblastoma multiforme, the two most aggressive tumors of the high-grade glioma group. Perfusion MRI studies report no significant difference in rCBV for both of them (Covarrubias, Rosen et al. 2004).

#### **4.4 Assessment of permeability with perfusion MRI**

A link between permeability, angiogenesis and malignancy in gliomas is normally assumed. Several perfusion MRI studies suggest a correlation between permeability and malignancy, but they do not establish a link between angiogenesis and permeability or angiogenesis and malignancy. What is widely accepted is that high permeability to macromolecules, due to the presence of endothelial gaps in tumor vasculature, is an important mediator of tumor growth and a pathognomonic feature for malignancy. Permeability studies are typically based in the determination of the transfer constant of contrast between the plasma and the extravascular extracellular space (Ktrans). Measurements of Ktrans are increasingly popular as a method for the quantification of contrast enhancement since they are designed to describe the distribution of contrast agent, free from scanning and machine dependant variables, which allow comparison of results from different studies and imaging centers. Nowadays, perfusion MRI parameters such as Ktrans and CBV can be directly correlated with histopathological changes in the brain.

T1W dynamic contrast MRI is probably the preferred method to analyze vascular permeability in gliomas and for mapping tumor angiogenesis. There are important studies showing a strong correlation between tumor grade and microvascular permeability by using DCE imaging methods (Roberts, Roberts et al. 2002). The net result is that blood vessel

high-grade gliomas, metastases, and lymphomas. These methods assign the most aggressive and malignant grade to the most enhanced tumors, but regarding the gliomas, this is not always reliable. A high-grade glioma may be often mistaken for a low-grade one when images show minimal edema, no gadolinium enhancement, no necrosis and no mass effect. Equally, low-grade gliomas can depict sometimes peritumoral edema, contrast enhancement, necrosis and mass effect leading to a high-grade false diagnosis (Law, Yang et al. 2003). In histopathology grading of gliomas, tumor cellularity and vascularity are the major factors to be taken into account. This information is accessible by the advanced functional MRI methodologies detailed in this chapter and can help in brain tumor grading. As regarding perfusion-based maps it is widely accepted that high regional CBV is more likely in high-grade gliomas. This parameter can reflect tumor vascularity and be used as an indicator of histological grade. In fact, many groups have shown the ability of DSC MRI for estimating the histopathological glioma grade. Several studies reported rCBV values measured in a large number of brain tumor patients suggesting that this parameter can improve distinction between high-grade and lowgrade gliomas (Aronen, Gazit et al. 1994; Calli, Kitis et al. 2006) with a high true-positive rate and a very low false-negative rate. On the other hand, gliomas have a great histopathological variability and in this sense, perfusion maps allow also for the localization of foci with very high rCBV values within the tumor. PWI can thus be very helpful not only for a valuable preoperative grading, but also for guiding the surgeon to the most appropriate site to obtain a representative biopsy of the most malignant portion. Although CBV is the most commonly used hemodynamic parameter in glioma grading, the largest limitation, entailing special clinical relevance, is the lack of discrimination between anaplastic gliomas and glioblastoma multiforme, the two most aggressive tumors of the high-grade glioma group. Perfusion MRI studies report no significant difference in rCBV for

A link between permeability, angiogenesis and malignancy in gliomas is normally assumed. Several perfusion MRI studies suggest a correlation between permeability and malignancy, but they do not establish a link between angiogenesis and permeability or angiogenesis and malignancy. What is widely accepted is that high permeability to macromolecules, due to the presence of endothelial gaps in tumor vasculature, is an important mediator of tumor growth and a pathognomonic feature for malignancy. Permeability studies are typically based in the determination of the transfer constant of contrast between the plasma and the extravascular extracellular space (Ktrans). Measurements of Ktrans are increasingly popular as a method for the quantification of contrast enhancement since they are designed to describe the distribution of contrast agent, free from scanning and machine dependant variables, which allow comparison of results from different studies and imaging centers. Nowadays, perfusion MRI parameters such as Ktrans and CBV can be directly correlated with

T1W dynamic contrast MRI is probably the preferred method to analyze vascular permeability in gliomas and for mapping tumor angiogenesis. There are important studies showing a strong correlation between tumor grade and microvascular permeability by using DCE imaging methods (Roberts, Roberts et al. 2002). The net result is that blood vessel

both of them (Covarrubias, Rosen et al. 2004).

histopathological changes in the brain.

**4.4 Assessment of permeability with perfusion MRI** 

permeability in gliomas can be used as a non invasive surrogate marker able to provide information related to rate of angiogenesis in the tumor. High-grade gliomas depict high vascular permeability, a consequence of the massive BBB disruption leading to extravasation of contrast material into the extravascular space that becomes detectable in MR images.

Although studies in humans have been typically carried out with DCE methods, it is also possible to perform and compare permeability measurements in grading gliomas employing T2\* DSC techniques, within certain limitations (Provenzale, Wang et al. 2002). In this case, the Ktrans maps can be generated by employing a calculation that allows the separation of T1 effects from the analysis. The advantage of using this method is that it is possible to obtain CBV and permeability maps from the same data set, providing simultaneously, two types of valuable information.

#### **4.5 Assessment of treatment response with perfusion MRI**

At present, the predominant method to assess tumor response to treatment is to measure the size of the lesion by using conventional, paramagnetically enhanced MRI. However, tumor size is not always specific of tumor behaviour and neoplastic features associated with aggressiveness of the tumor may vary without changes in the tumor size. On the other hand, since angiogenesis plays a critical role not only in the development of brain tumors but also in the malignant transformation, many of the novel therapies currently investigated are focused on antiangiogenic strategies. Monitoring the success of such therapeutic methods relies on the ability to quantify tumor vascularization through the measurements of hemodynamic parameters. It is widely accepted now that the use of quantitative measurements of tissue enhancement characteristics provides a more valuable indicator to monitor the therapeutic response of tumours to novel antiangiogenic drugs in the treatment of gliomas. For example, rCBV is supposed to reflect microvessel density, an early response in this parameter to antiangiogenic therapies should show a decrease, thus providing the potential to predict growth or regression.

Figure 13 shows a clear example of this behaviour. While T1W images show tumor growth even 6 weeks after initiating the treatment, rCBV maps from PWI depict a much earlier drop in tumor perfusion. This demonstrates that tumor size and perfusion may lead opposite conclusions because they represent indeed, different tumor features.

Because the increase in endothelial permeability is a primary effect of vascular endothelial growth factor (VEGF) and other angiogenic cytokines, changes in the capillary leakage of contrast media may offer a predictable response to anti-angiogenic therapies. Quantification of the enhancement effect is therefore a good candidate as a potential biomarker of drug activity. Changes in permeability occur rapidly in response to VEGF inhibitions, so assessment of these changes either by DSC or DCE can be used as a marker of antiangiogenic therapy responses. Although quantification of permeability would be an optimal strategy, this is in fact, extremely difficult to achieve. Along these lines, several pharmacokinetic models to analyze dynamic enhancement data have been developed to estimate Ktrans and used it as an therapeutic indicator (Jackson, Jayson et al. 2003). Now, it is accepted and recommended the use of perfusion MRI to monitor tumor angiogenesis and evaluate the response to antiangiogenic therapies, a subject that has gained considerable clinical and research efforts, even leading to published recommendations concerning this important capacity of MRI (Brasch, Li et al. 2000).

This work was supported in part by grants CTQ2010-20960-C02-02 to P.L.L. and grant SAF2008-01327 to S.C. A.M.M. held an Erasmus Fellowship from Coimbra University and E.C.C. a predoctoral CSIC contract. The valuable contribution of Mr. Javier Perez drafting

Aronen, H. J., I. E. Gazit, et al. (1994). "Cerebral blood volume maps of gliomas: comparison

Aronen, H. J. and J. Perkio (2002). "Dynamic susceptibility contrast MRI of gliomas."

Basser, P. J. and D. K. Jones (2002). "Diffusion-tensor MRI: theory, experimental design and

Brasch, R. C., K. C. Li, et al. (2000). "In vivo monitoring of tumor angiogenesis with MR

Brunberg, J. A., T. L. Chenevert, et al. (1995). "In vivo MR determination of water diffusion

gliomas of the cerebral hemispheres." AJNR Am J Neuroradiol 16(2): 361-71. Bulakbasi, N., M. Kocaoglu, et al. (2003). "Combination of single-voxel proton MR

Buxton, R. B. (2005). "Quantifying CBF with arterial spin labeling." J Magn Reson Imaging

Buxton, R. B., L. R. Frank, et al. (1998). "A general kinetic model for quantitative perfusion imaging with arterial spin labeling." Magn Reson Med 40(3): 383-96. Calli, C., O. Kitis, et al. (2006). "Perfusion and diffusion MR imaging in enhancing malignant

Covarrubias, D. J., B. R. Rosen, et al. (2004). "Dynamic magnetic resonance perfusion

Chenevert, T. L., L. D. Stegman, et al. (2000). "Diffusion magnetic resonance imaging: an

Chien, D., R. B. Buxton, et al. (1990). "MR diffusion imaging of the human brain." J Comput

Detre, J. A., J. Wang, et al. (2009). "Arterial spin-labeled perfusion MRI in basic and clinical

Ellingson, B. M., M. G. Malkin, et al. "Validation of functional diffusion maps (fDMs) as a biomarker for human glioma cellularity." J Magn Reson Imaging 31(3): 538-48. Evelhoch, J. L. (1999). "Key factors in the acquisition of contrast kinetic data for oncology." J

Folkman, J. (1971). "Tumor angiogenesis: therapeutic implications." N Engl J Med 285(21):

Furnari, F. B., T. Fenton, et al. (2007). "Malignant astrocytic glioma: genetics, biology, and

early surrogate marker of therapeutic efficacy in brain tumors." J Natl Cancer Inst

coefficients and diffusion anisotropy: correlation with structural alteration in

spectroscopy and apparent diffusion coefficient calculation in the evaluation of

with tumor grade and histologic findings." Radiology 191(1): 41-51.

data analysis - a technical review." NMR Biomed 15(7-8): 456-67.

common brain tumors." AJNR Am J Neuroradiol 24(2): 225-33.

**6. Acknowledgements** 

22(6): 723-6.

92(24): 2029-36.

1182-6.

Assist Tomogr 14(4): 514-20.

Magn Reson Imaging 10(3): 254-9.

**7. References** 

the illustrations is gratefully acknowledged.

Neuroimaging Clin N Am 12(4): 501-23.

imaging." Acad Radiol 7(10): 812-23.

cerebral tumors." Eur J Radiol 58(3): 394-403.

imaging of brain tumors." Oncologist 9(5): 528-37.

neuroscience." Curr Opin Neurol 22(4): 348-55.

paths to treatment." Genes Dev 21(21): 2683-710.

Fig. 13. Change in perfusion in a pathologically confirmed GMB case before (A) and after (B) the initiation of anti-angiogenic chemotherapy. Left images in both panels show contrast enhanced T1W images used to measure tumor volume (green). Right images depict CA leakage corrected rCBV maps showing tumor perfusion (red arrow) (Sawlani, Raizer et al.) Reproduced with permission.
