Quantitative MRI in Cancer (Imaging in Medical Diagnosis and Therapy)
Book file PDF easily for everyone and every device.
You can download and read online Quantitative MRI in Cancer (Imaging in Medical Diagnosis and Therapy) file PDF Book only if you are registered here.
And also you can download or read online all Book PDF file that related with Quantitative MRI in Cancer (Imaging in Medical Diagnosis and Therapy) book.
Happy reading Quantitative MRI in Cancer (Imaging in Medical Diagnosis and Therapy) Bookeveryone.
Download file Free Book PDF Quantitative MRI in Cancer (Imaging in Medical Diagnosis and Therapy) at Complete PDF Library.
This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats.
Here is The CompletePDF Book Library.
It's free to register here to get Book file PDF Quantitative MRI in Cancer (Imaging in Medical Diagnosis and Therapy) Pocket Guide.
Figure 4. Liver metastases in a year-old female patient presenting with colon adenocarcinoma, undergoing treatment with antiangiogenic drug bevacizumab. The necrotic center of the metastatic lesions green ROI and white arrow shows attenuation of signal intensity, with increasing b values indicating less restriction to diffusion. In comparison, the peripheral zone of the tumor purple ROI presents increased cellularity and little signal attenuation with the increase of the b value.
Applying different b values it is also possible to perform a quantitative analysis with DWI. This particular analysis is usually performed on a workstation by calculating ADC values Figure 5. The ADC is calculated for each image pixel and is shown as a statistical parametric map. Areas with restriction to water molecules diffusion demonstrate increase signal intensity at DWI and low values at the ADC mapping: demonstrating the correspondence of both Figure 6. Figure 5. Quantitative analysis of DWI. The anatomopathological analysis confirmed the tumor heterogeneity, a moderately differentiated cholangiocarcinoma with areas of necrosis.
Figure 6. On A , axial post-gadolinium T1-weighted image showing two ROIs: 1 purple located in the tumor lesion, and 2 green located in the healthy tissue. On C , the rBV map superimposed on the contrast-enhanced axial T1-weighted image demonstrates increased perfusion in the tumor ROI arrow.
A great part of the molecular imaging techniques utilize exogenous markers that produce the signal from the particle itself or from the pattern of contrast enhancement. For example,, the marker may be a conventional intravenous contrast medium such as gadolinium. The utilization of such agents has occurred with the advent of the dynamic contrast enhanced DCE technique or perfusion MRI.
In such techniques, the images are sequentially acquired during the contrast agent passage through the tissue of interest, allowing the characterization of lesions in different anatomical sites, including brain, breast, gynecologic and prostate lesions. Such methods are not intrinsically molecular, but allow for an indirect evaluation of molecular processes that affect the blood flow The quantitative and qualitative dynamic analyses of the MRI contrast enhancement may also be useful in the differentiation of benign from malignant musculoskeletal system tumors Nowadays a very promising perfusion MRI technique without the utilization of paramagnetic contrast is available.
Such a technique, called arterial spin labeling, has been utilized in the evaluation of the cerebral blood flow, but currently it is available only in more advanced centers The conventional dynamic MRI technique is based on the concept of development of new vessels angiogenesis associated with increased blood flow and vessels permeability, which constitute essential conditions for metastatic dissemination of malignant tumors 14 , 16 , The microvascular structure of the tumor constitutes a relevant prognostic factor, and perfusion MRI can provide information about this special characteristic in a noninvasive way The images are often acquired after infusion of low molecular weight gadolinium.
The distribution of the contrast medium in the intra- and extravascular regions will depend upon some factors such as blood flow, vascular permeability and interstitial diffusion capacity. The GRE-T1 sequences can more appropriately characterize the alterations in vessels permeability and extravasation to the extravascular space and are indicated for the evaluation of extracranial regions.
With those sequences it is possible to perform qualitative, semi-quantitative and quantitative measurements 13 , Qualitative measurements can be obtained by means of signal intensity-time curves, often utilized in the evaluation of breast carcinomas Figure 7. Semiquantitative measurements are related to the differences in signal intensity before and after contrast medium infusion relative signal intensity. Quantitative measurements are based on pharmacokinetic models which allow for data collection.
Thus, it is possible to create color parametric maps demonstrating the tumor behavior which is important, for example, for the therapeutic planning 13 , 14 , 16 Figure 6. Figure 7. A year-old female patient presenting with multicentric invasive ductal carcinoma in the right breast and fibroadenoma in the left breast. On A, one observes contrast-enhanced, dynamic 3D MIP image with subtraction, demonstrating the presence of multiple breast nodules at right and one retroareolar nodule at left.
On B , C and D it is possible to observe that the contrast-enhancement pattern of the breast nodules at right demonstrates characteristics of washout curve and plateau type 3, sometimes observed in malignant nodules. On E , the analysis of the kinetic curve of the breast nodule at left demonstrates characteristics different from the others, showing a type 1 curve, a pattern that is more frequently observed in benign nodules. The data provided by dynamic MRI may be utilized for different purposes.
Such evaluation may occur in different phases including diagnosis, staging and treatment response evaluation, particularly in cases where antiangiogenic drugs need to be evaluated 14 , Data in the literature also attribute a relevant role of perfusion MRI as a prognostic factor and in the evaluation of disease recurrence. Perfusion MRI has been utilized in the evaluation of different types of tumors as a tumor hypoxia biomarker, particularly in cases of well vascularized tumors, such as those in the lungs, uterine cervix, head and neck, breast, liver, musculoskeletal system and colorectal tumors 14 , 16 , However, the presence of prominent contrast enhancement at the end of treatment may be associated with a locally aggressive disease, with reduction of survival rates.
Angiogenesis studies have continuously evolved over the last years. The advent of macromolecular contrast agents allows for the maintenance of such agents for longer periods in the intravascular spaces. Contrast media containing gadoxetic acid are examples of such agents in the characterization of focal liver lesions. The development of substances directed against molecules expressed by neoangiogenic vessels, as the factor of endothelial vascular growth, is another application field by the perfusion technique 16 , Magnetic resonance spectroscopy MRS evaluates the distribution and levels of metabolites normally found in healthy tissues as well as increased levels of metabolites usually detected in within tumor Creatine, choline, lactate, citrate, N-acetyl aspartate and adenosine triphosphate are examples of altered metabolites which are commonly found This technique can be indicated, for example, to evaluate breast, prostate and brain lesions 18 — The main indications of this method are the following: lesion characterization, selection of biopsy site, and evaluation of therapeutic response, among others.
This technique can be applied in the evaluation of brain lesions. Increased levels of choline considered a marker for cell proliferation in association with decreased levels of creatine considered a marker for energetic processes and decreased levels of N-acetyl aspartate considered a neuronal marker have been found in the evaluation of brain neoplasms. Combining such levels, it is possible to differentiate, for example, low-grade from high-grade gliomas 18 , 19 Figure 8. It can differentiate viable tumor from necrotic area important in the evaluation of the tumor response.
In the presence of response, a decrease in the choline and N-acetyl aspartate peaks is observed in association with increased levels of lipids and lactate anaerobic markers Figure 8. A year-old male patient. Tumor resection glioblastoma multiforme in the left temporal lobe six months ago, undergoing treatment with radiotherapy and temodal. MRI scan with advanced techniques was requested for differential diagnosis between recurrence and radionecrosis in post-gadolinium enhancement areas in the surgical site. On A, one observes contrast-enhanced T1-weighted image demonstrating enhancement of the surgical site arrow.
On C and D , the spectroscopy study demonstrates decreased peak of the metabolite N-acetyl aspartate NAA and increased choline peak Cho , corroborating the diagnosis. In the evaluation of breast lesions, for example, association with choline peak may be detected in malignant lesions. However, in benign lesions or in healthy breast tissues, choline levels are either low or undetectable.
In the evaluation of the prostate spectroscopy obtains metabolic data based on the relative concentration of endogenous metabolites such as choline, creatine, citrate and, most recently, polyamine Thus, spectroscopy may be employed in the diagnosis of tumor recurrence, in patients treated by radiotherapy, cryotherapy or surgery 21 — The routine utilization of spectroscopy in the evaluation of other neoplasms is still questionable Figure 9.
Figure 9. On B , T2-weighted image demonstrates a subtle ill defined area arrow. On D , the ADC map demonstrates the same area with low signal intensity arrow at 6 o'clock in the peripheral zone. On E , the dynamic contrast-enhanced image demonstrates enhancement in the medial region from 5 o'clock to 7 o'clock. On F , the kinetic curve presents intense and early enhancement washin tending towards rapid clearance washout. Such parameters represent an area suspected for malignancy. Whole-body imaging modalities have been utilized for some time in the evaluation of cancer patients However, the development of new MRI sequences has been improving the utilization of the method in the evaluation of cancer patients The introduction of echo-planar techniques has allowed the acquisition of whole-body images by means of different sequences such as T1-weighted, T2-weighted, STIR and diffusion.
The better management of the effects from artifacts generated by physiological cardiac and respiratory motion has allowed for the acquisition of good functional images which supplement morphological data usually obtained by conventional MRI techniques Figure 10 Figure WBMRI is useful in the detection of metastases, particularly in brain, liver and bone lesions. On this figure, a year-old female patient presenting with lung adenocarcinoma in the upper right lobe long arrow with metastasis to the left adrenal gland short arrow. Whole-body MRI WBMRI is a noninvasive technique free from the risks of ionizing radiation and with high resolution for soft tissues, which can rapidly acquire whole-body images.
During the scanning, the body is divided into different portions, and the images are acquired in axial and coronal sections Diffusion-weighted whole-body imaging may be applied to obtain images with body signal suppression. Histology is typically carried out on biopsy samples, which provide only focal information on a heterogeneous mass. Sample collection constitutes a burden for the patient and may not always be feasible. Furthermore, longitudinal analyses are difficult. On the other hand, histology yields unambiguous information critical for diagnosis that is based on cellular morphology or on the expression of a characteristic molecular signature expressed by the tissue.
The possibility to simultaneously analyze multiple tissue parameters is essential for the identification of the tumor type. Non-invasive imaging for tumor diagnosis offers unique advantages: minimal burden of the patient, full three-dimensional sampling of the heterogeneous lesion, dynamic measurement of physiological and metabolic processes complementing morphological information, and the possibility for longitudinal examinations.
Yet, current imaging approaches are based on structural and physiological phenotypic readouts, which are sufficient for lesion detection and monitoring disease progression or therapy response, but most likely, will not allow identifying the lesion type. Analogous to histological tissue characterization it would be important to assess a molecular and cellular characteristics and b multiple complementary tissue features in order to achieve a high discriminative power.
As we will see later, the use of complementary imaging modalities that probe different aspects of the pathology would be most promising. Nevertheless, we will focus our current discussion on magnetic resonance based techniques, which are attractive as they provide high spatial resolution, unique soft tissue contrast, a temporal resolution sufficient for studying dynamic processes, and moreover are characterized by high chemical specificity, a feature that is extensively used for chemical and biochemical structure elucidation.
In addition, the method can be easily translated into the clinics. Magnetic resonance images represent a weighted distribution of protons 1 H in tissue, the predominant source of the signal being tissue water and lipids adipose tissue. Obviously the signal is proportional to the density of protons in the respective tissue. The weighting function is governed by the proton magnetic properties, which are affected by their local environments due to magnetic and chemical interactions which depend on the nature of tissue Weishaupt et al.
- Improving prostate cancer diagnosis with MRI 'risk maps' - UChicago Medicine.
- Wildlife Forensic Investigation: Principles and Practice.
- Cancer Imaging.
- Services on Demand?
The effect of the environment on the MRI signal is lumped into parameters describing three distinct relaxation processes Mark Haacke, : 1 the longitudinal relaxation characterized by the relaxation time T1, which describes the interaction of the spin with its environment, hence the expression spin-lattice relaxation as a crystal lattice constituted the environment in early solid state physics nuclear magnetic resonance NMR experiments.
T1 relaxation is based on energy exchange between the spin under investigation and its environment and occurs such that the system is driven back to its thermal equilibrium state. T2 relaxation is based on dipole—dipole interactions between spin pairs that fluctuate with regard to their spatial alignment and hence is of stochastic nature. It leads to the irreversible loss of phase coherence and hence to a loss in signal intensity.
This local field inhomogeneities are static and hence deterministic and can be accounted for when tailoring the MRI data acquisition so-called spin-echo experiments. Relaxation processes can be influenced by administration of contrast agent, which are either paramagnetic gadolinium based or superparamagnetic agents iron-oxide based. These agents contained unpaired electrons with a strong effect on the local magnetic field that is experience be nearby protons. The contrast mechanism of the two classes of agents is different, yet a detailed description is beyond the scope of this article Rudin, a.
In the context of our discussion it suffices to state that paramagnetic agents enhance the longitudinal relaxation rate, i. Apart from enhancing the contrast in static MR images to improve discrimination of distinct tissues, MRI allows monitoring dynamic changes following the contrast agent administration. The contrast change measured in a volume element voxel is proportional to the amount of contrast agent in this voxel, which by itself depends on the biodistribution including compartments within a tissue and pharmacokinetic properties of the agent. Such dynamic studies yield information on tissue perfusion, vascular leakage, or distribution volumes.
The magnetic resonance phenomena are not only restricted to the detection of protons of water and lipid molecules in tissue. Essentially all magnetic nuclei give rise to signal. The resonance frequency of a nucleus depends on its identity characterized by the so-called gyromagnetic ratio and its chemical environment. It is in particular the fact that the magnetic resonance sensitively probes the chemical structure to which the interrogated nucleus is attached that has made the method indispensable for chemical structure elucidation.
The identification of a molecular entity is based on the detailed spectral analysis of its resonance frequencies. Translating these approaches to in vivo tissue characterization therefore bears considerable potential to enable a detailed molecular tissue characterization, which might be of high diagnostic value. Apart from protons, other nuclei such as phosphorus, carbon, constituents of many biologically relevant molecules are of interest for in vivo magnetic resonance spectroscopy MRS.
Yet this method suffers from the low intrinsic sensitivity of magnetic resonance, as these metabolites are typically present at millimolar to sub-millimolar concentration compared to water protons with tissue levels of approximately 80 M. If compared to healthy organs, tumor tissues present in general highly heterogeneous and chaotic architecture. Such heterogeneity is primarily due to the genetic instability of tumor cells that is responsible of the apparently chaotic tumor development, which is reflected in tissue architecture, tumor vasculature, host infiltrates, and metastasis formation Heppner, ; Marusyk et al.
This chaotic behavior occurs at a molecular, cellular, and microdomain level and determines also the interaction with the host environment. The result is the formation of different regions inside the tumor, which may exhibit completely different physiological behavior Denysenko et al. In order to rationalize the complexities of neoplastic disease, Hanahan and Weinberg have defined six phenotypic hallmarks of cancer, which correspond to six biological features acquired during tumor development.
Those include sustained proliferative signaling, evasion of effects of growth suppressor, resistance to cell death program, acquisition of replicative immortality, development of a vascular network angiogenesis , invasion of adjacent healthy tissue, and the formation of distant metastases. In a recent publication Hanahan and Weinberg, , these initial six hallmarks were complemented by four additional features related to the specific behavior of tumor tissue: genome instability, inflammation, reprogramming of energy metabolism, and evasion of immune surveillance.
An important aspect of tumor is that they are not only composed of cancer cells but contain a variety of host derived cells such as immune cells, endothelial cells, pericytes, fibroblasts, stem, and progenitor cells that characterize the hallmarks traits and constitute the tumor microenvironment Swartz et al. Considerable efforts have been invested to assess these tumor hallmarks non-invasively using imaging. Yet, all these phenotypic readouts are not specific enough for an unambiguous identification of the tumor type, which is based on unique molecular markers.
Secondly, many of these tools are still in an early experimental stage and will not be available in a clinical setting soon. Damadian reported on the observation that T1 relaxation times in tumors are higher than in the adjacent normal tissue and suggested that this feature might be used for tumor detection. This constituted one of the prime motivations that later led to the development of MRI. Nowadays, modern MRI scanners offer several tools for detecting and characterize tumor.
Despite the fact that the basic biophysical mechanism leading to tissue specific relaxivity values are poorly understood, the evaluation of relaxivity parameters are of high diagnostic value. According to the type of MR sequence and the relative parameters, it is possible to acquire a signal, which is mostly dominated by one of these contributions. Most established are T1-weighted, T2-weighted or proton density weighted images Haacke et al. By optimizing the contrast between neoplastic and normal tissue it is in generally possible to detect the cancer lesion, to identify sub-regions displaying different tissue characteristics dense versus non-dense tissue, poorly versus highly vascularized, necrotic areas, edematous tissue, etc.
Instead some generic tissue features are reflected. For example, T1-weighted images are usually used to assess the gross morphology of the tumor as shown in Figure 1 left. As rule of thumb, regions with high water content appear dark, while regions with high fat content appear bright Weishaupt et al. In combination with gadolinium-based contrast agent such as Gd-DTPA it is possible to assess regions displaying high uptake of the agent indicative of hemorrhage and leaky vessels.
Areas, for which little uptake is observed are commonly associated with necrotic or edematous domains. Only when waiting sufficiently long these areas will accumulate extravasated contrast agent via passive diffusion. T1-weighted image of a glioma following contrast enhancement using a gadolinium-based contrast agent left. Adapted from Young , reproduced with permission.
In T2-weighted images areas with high water content appears bright. Since most diseases are characterized by increased water content in tissues associated with an inflammatory tissue response, T2-weighted are particularly useful for pathological investigation. Dark regions may indicate high blood content such as hemorrhage, vessels, or angiomas.
In proton weighted images Westbrook, , bright areas indicates high proton density tissue, such as cerebrospinal fluid or edema, while dark areas indicate low proton density such connective tissue i. Nowadays, tumor detection based on altered T1 and T2 relaxivity values is commonly used to diagnose and follow-up different kinds of tumor comprising, among the others, brain tumor Young, , breast tumor Heywang-Kobrunner et al.
By means of T1 and T2 weighted images and in combination with contrast agent, as Gd-DTPA or superparamagnetic nanoparticles, it is possible to assess tumor morphology and grossly identify edematous and necrotic regions. Moreover, kinetics and extent of contrast agent uptake are considered as an indicator of prognostic quality. The possibility to obtain high-resolution and high-contrast images of soft tissue with similar density but different relaxivity values makes MRI the method of choice for the detection of solid tumors.
Diffusion Weighted Imaging DWI measures the random movement of the water molecules and allows deriving the so-called apparent diffusion coefficient ADC for each voxel Haacke et al. Also, structural barriers like cell membranes, or perfusion effects affect diffusion Haacke et al. Hence, regions with densely packed cells will show low ADC values. This has been exploited in the characterization of brain neoplasms.
High grade tumor neoplasms display significant reduction of ADC and correspondingly a higher signal in DWI as compared to lower grade Okamoto et al. Fluid filled cysts or edematous regions appear hyperintense in ADC maps and hypo-intense in DWI when compared to the normal parenchyma because they largely correspond to bulk water enabling unrestricted diffusion within the MRI timescale; Drevelegas and Papanikolaou, Recent data have expanded the concept that inflammation is a critical component of tumor progression Coussens and Werb, The quantification of the inflammatory status is crucial in the determination of the tumor volume, since its value is an important prognostic factor with regard to the treatment of malignant tumors Xie et al.
Moreover, inflammation may also influence therapy outcome in two opposite ways, in particular for brain tumors such as gliomas Kleijn et al. It can lead to tumor control, by killing cancer cells and establishing anti-cancer immunity, or it may further promote tumor growth, by participating in glioma reoccurrence and progression. It is therefore evident that the possibility to monitor the inflammation status in vivo , i. Traditionally, such evaluation is performed ex vivo using cytometry and immunohistochemistry methods, or in vivo using labeled-radionuclides for PET Positron Emission Tomography or SPET Single Photon emission tomography scanner Ahrens and Bulte, However, recent developments, in particular the possibility to prepare non-toxic MRI probes for cell labeling, enables MRI based tracking of immune cells.
Immune cells can be labeled with superparamagnetic iron oxide based SPIO nanoparticles in two ways: i by ex vivo labeling of harvested cells that are incubated with SPIO nanoparticles in media typically using a transfection agent, or ii by non-selective in situ labeling of the phagocytic cells, such as macrophages, following intravenous injection of SPIO nanoparticles Bhakoo et al. PFC emulsion can be used to track cells using the same labeling strategies.
PFC-based cell tracking provides high specificity for cell detection i. Yet they require a specific MRI coil tuned to the resonance frequency of 19 F nuclei. Disadvantages of using passive labeling strategies are that only the presence of the label is detected, which is not necessarily identical with the presence of cells. Cells may release the label into the environment, e. Also, the presence of the label does not yield any information on the status of the cell, i.
Finally, for dividing cells which is not relevant for the immune cells the label will be subsequently diluted. In addition, a passive label will be degraded over time. Genetic encoded reporters avoid some of these issues. They only yield a signal when the gene is expressed, i. On the other hand, the sensitivity of genetic cell marking is in general inferior to that of potent exogenous labels.
A Imaging of in vivo antigen capture and trafficking of dendritic cells DCs. Sentinel DCs were labeled in situ by intradermal injection of unlabeled dashed arrow or SPIO-labeled solid arrow irradiated cancer cells, which function as a vaccine. Following phagocytosis of both SPIO particles and tumour antigens in a process known as magnetovaccination, the hypointense DCs migrate into the medulla of the draining popliteal lymph node.
Numerous bright spots PFC droplets are observed inside the cell. Particles appear as smooth spheroids Ogawa et al. Arrowheads indicate vesicles. The scale bar represents nm. Adapted from Ahrens and Bulte , reproduced with permission. Magnetic resonance imaging cell tracking can also be used to monitor inflammation related to other disease as neurological disorders, autoimmune diseases, or transplant rejection.
Moreover, it is likely to become an important tool also in cell therapy i. One of the consequences of the inflammatory status is the formation of a peritumoral edema which is the results of several cellular mechanism Stummer, Although its prognostic value for diagnosis, as well its role in the course of disease is still a matter of discussion, peritumoral edema may cause severe neurological symptoms in case of brain tumor, and remains a challenge in the treatment of glioblastoma patients Kleijn et al.
The evaluation of edema by means of MRI is usually performed using T2-weighted sequences that are quite sensitive to water content, and by assessing changes in ADC. The regions affected by edema are characterized by prolonged T2 values and therefore appear hyperintense in T2-weighted images. The physiology of tumor tissues is directly dependent on the structure and functionality of the vascular network developed during tumor growth.
The newly formed vessels are responsible for the delivery of the nutrients from the hosting tissue to the tumor and for the removing of waste metabolites from the tumor. Tumor vasculature deviates profoundly from that of the normal organs both in vascular architecture and functionality. The vascular network of solid tumor does not show the hierarchical branching patterns characteristic for the majority of healthy organs. This is the results of the opportunistic nature of the angiogenic process, which in tumor seems not to follow physiological pre-determined steps Tropres et al.
Initially avascular tumor masses trigger the development of new angiogenic vessels as a consequence of hypoxia and the secretion of angiogenic factors Lemasson et al. Alternatively, tumors may grow along one or more existing vessels and co-opt them in the tumor structure in a parasitic manner. In both cases vessels usually remain in a primitive status with immature vascular walls and proper support by the tissue matrix. Tumor vascular networks therefore consist of tortuous micro-vessels exerting chaotic branching, arterial-venous shunts, and are subject to acute or transient collapse Heywang-Kobrunner et al.
The lack of maturation of the primitive vessel network gives origin to a few abnormalities in vascular function. Tumor capillaries show high permeability compared to the healthy ones Tropres et al. This results in a profound extravasation of erythrocytes and plasma in the adjacent tissue leading to an elevated interstitial fluid pressure and to a rise in the viscous resistance to blood flow Dominietto, Second, because of this resistance and chaotic structure, the blood circulation or perfusion within such vessels is rarely correlated to the metabolic demands of solid tumor Heywang-Kobrunner et al.
Moreover, the clearance of metabolites from the tissue and the drainage by the venous system do not work properly and are responsible of the accumulation of blood in the tumor tissue. To complicate matters even more, the degree of abnormalities changes in different kinds of tumors and also during different stages of the same tumor. While from a biological point of view the origin of these physiological fluctuations is poorly understood, the assessment of vascular abnormalities constitute an attractive biomarker, as it clearly distinguishes neoplastic from normal tissue.
Various structural and physiological aspects of tumor vasculature can be quantified by MRI and used for classification and staging of tumors. All these approaches aim at generating a high contrast between the vascular lumen blood compartment and the surrounding tissue to enable the segmentation and extraction of vascular structures. Magnetic resonance angiography of a brain tumor to evaluate the tortuosity of the vascular network. Vessels within the tumor nidus are shown in red, vessels supplying or passing through the nidus in gold, while normal vessels outside the nidus are blue.
The nidus, containing type II tortuosity vessels, is volume rendered at full opacity left , at partial opacity center , while vascular structures exclusively are shown at right. Adapted from Bullitt and Gerig , reproduced with permission. Time-of-flight angiography Heverhagen et al. Briefly, by a combination of radiofrequency excitation pulses all the spins of the excited volume will be saturated and, because of that, the signal will be largely suppressed. However, blood that has entered the imaged volume, will give rise to the full signal intensity, as it has not experienced previous saturation.
Contrast enhanced Chandra et al. Gadolinium based contrast agent will produce an enhancement of the signal in T1-weighted sequences, while iron-based contrast agent will cause dephasing of the nuclear magnets decreasing the overall signal in T2-weighted acquisitions.
Data acquisition and preprocessing
Acquisition has to be fast enough that extravasation of the contrast agent remains minimal. Angiograms are then obtained by comparing pre- and post-contrast images. Phase contrast Thomas and Wells, utilizes the change in the phase shifts of the flowing protons in the region of interest to create an image. Spins moving along the direction of a magnetic field gradient receive a phase shift proportional to their velocity. This is usually accomplished by applying gradient pairs, which sequentially dephase and then rephase spins during the sequence.
Use of phase-sensitive image reconstruction allows depticting the vascular systems exclusively and more over provides information on blood flow velocities. Vessel size imaging Tropres et al. From indirect measurements of vessel surface and volume we can infer on the average radius of the vessels in a given region-of-interest. The dimension and density of the vessels is an important index when studying angiogenesis. When combined with an independent measurement of the tumor blood volume TBV , it constitutes an index of the organization of the vascular network.
Identification of vessels of various diameter from big to small indicates a hierarchical network, while the presence of only small vessels is an index of the poor organization of the vascular tree. While information on the vascular architecture within the tumor is a downstream manifestation of the angiogenic process, it is important to derive physiological information in order to understand the implication on substrate delivery, which essentially determines the fate of the tumor.
Capillary vessels like arterioles and venules are permeable to the substances present in the blood to enhancing compound exchange between the blood and tissue compartment. It has been shown that in tumors also relatively big vessels are highly permeable due to the immature structure of the vascular wall.
This results on an almost completely leaky network with a highly non-uniform blood supply to tumor tissue Dominietto, The characteristically high permeability of tumor vessels has been suggested as biomarker for angiogenesis Feng et al. Vascular permeability values are commonly assessed by means of T1-weighted dynamic contrast enhanced DCE acquisitions, involving serial images of the same region during the administration of a gadolinium-based contrast agent Rudin et al.
The measured MRI signal enhancement curve is fitted using a two-compartment model originally proposed by Tofts and Kermode In its simplest version the model comprises a vascular and an extracellular compartment. Fitting to the enhancement curve is carried out by optimizing two parameters, the vascular permeability defined by the transfer constant k trans , a measure for the rate of contrast agent extravasation, and the volume of the extracellular compartment V e.
While TBV measure the volume of the vascular compartment in a region-of-interest, TBF assess the exchange of blood within this volume per unit time. For data analysis, it is assumed that, due to its nanoparticulate size, the contrast agent remains confined to the blood compartment, at least for the duration of the measurement. The images show the effect of DMOG treatment that affects angiogenesis process left versus placebo right. The color bar indicates the rTBV values in arbitrary units. Adapted from Dominietto et al. The oxygenation is another important factor in tissue characterization since abnormal oxygen levels have several implications in tumor progression and treatment Nilesh and Quarles, In particular, a hypoxic environment is known to promote angiogenesis, inflammatory behavior, genetic instability, invasiveness, and metastasis formation.
Hence, hypoxia is associated with increased malignancy and causes reduced efficacy of radio- and chemo-therapy. This method has been used to monitor treatment response during phototherapy Gross et al. While BOLD based methods provide accurate qualitative information of blood oxygenation it is difficult to extract reliable quantitative data. Images show a signal enhancement maps color overlaid on T2-weighted anatomical images.
Images have been acquired 1 week before start of neoadjuvant chemotherapy left , after one cycle of chemotherapy showing small signal response middle and after four cycles of chemotherapy demonstrating a striking change in tumor characteristics in response to therapy right. Adapted from Jiang and Weatherall , reproduced with permission. It has been demonstrated that the 19 F relaxation time T1 is linearly dependent on oxygen tension Joseph et al. However, given the difficulty of delivering sufficient quantities of PFCs to tumor tissue, as many of these agents require intra-tumoral injection, the method has remained a preclinical tool Nilesh and Quarles, Metabolic reprogramming of tumor cells has been recognized already very early.
It has been observed that neoplastic tissue exerts high glycolytic activity even under conditions of normoxia Warburg effect; Gatenby and Gillies, In fact, measurement of enhanced glucose utilization with PET using [ 18 F]fluorodoxyglucose FDG as tracer has emerged as important diagnostic tool for tumor diagnosis, in particular for detection of the metastatic burden. Only recently, molecular mechanism underlying this reprogramming, linking metabolic processes to altered gene expression are being elucidated DeBerardinis et al. Glycolysis leads to the production of lactic acid from pyruvic acid via pyruvate dehydrogenase, which is responsible for acidosis.
Hence, despite increased acid production, tumor cells maintain a normal slightly alkaline intracellular pH. The major acid load is transported outside the cells but, since the acid cannot be easily removed by the abnormal vasculature, the microenvironment will become acidic Zhang et al. Tissue acidosis is an important feature of the tumor microenvironment which has been shown to drive local invasion and not surprisingly several approaches have been described to assess tumor pH non-invasively Figure 6.
MRS methods are generally based on a difference in chemical shifts between pH-dependent and pH-independent resonances Zhang et al. A resonance becomes pH dependent when the resonance frequency of the protonated form is distinct from that of the deprotonated form and when the exchange reaction is fast compared to the MRS time scale, which is defined by the frequency difference of the two resonances.
Different nuclei can be used to determine tissue pH using this approach: 31 P Gadian and Radda, , 1 H and hyperpolarized 13 C Gallagher et al. Adapted from Zhang et al. An alternative approach using MRI relies on perturbing the relaxivity of water via pH-dependent relaxation agents. Small molecules Gd-based agents, whose relaxivity is pH dependent, have been recently synthesized Zhang et al.
For the pH quantification, this method requires knowledge of the concentration of the agent in each voxel. Finally, a new generation of agents that have been developed to generate contrast via chemical shift saturation transfer CEST enable pH measurement Zhang et al. The dynamic process of CEST can be described by 2-pool chemical exchange model, wherein the magnetization is exchanged between a labile proton e.
The two resonances have to be distinguishable. In the experiment one of the two resonances the smaller proton pool is magnetically labeled saturated and the transfer of label to the exchange partner the water proton is monitored. For example, the resonance of amide protons is saturated and the transfer of saturation to the water resonance, i. Mathematical modeling based on Bloch equations coupled by chemical exchange yields estimates for the exchange rate, which depend on pH. In general, exchange rates are slower at a low pH.
The concentration various metabolites can be measured by means of MRS Figure 7. Seidel et al. Tolmachev, I. Velikyan, M. Ahlgren, H. Tran et al. Kinoshita, Y. Yoshioka, Y. Okita, N. Hashimoto, and T. Becker, C. Hessenius, K. Licha et al.
Molecular Imaging Probes for Diagnosis and Therapy Evaluation of Breast Cancer
Bugaj, S. Achilefu, R. Dorshow, and R. Smith-Jones, D. Solit, T. Akhurst, F. Afroze, N. Rosen, and S. Oude Munnink, M. Korte, W. Nagengast et al. Kramer-Marek, D. Kiesewetter, and J. Van De Ven, S. Elias, C. Chan et al. Wang, J. Yu, G. Jung, Y. Choe, J. Paik, and K. Ke, X. Wen, M. Gurfinkel et al. Hun and Z. Qi, Z. Miao, H.
International Journal of Biomedical Imaging
Liu, Y. Xu, Y. Feng, and Z. Heskamp, H. Van Laarhoven, J. Molkenboer-Kuenen et al. Heskamp, O. Boerman, W. Oyen, J. Molkenboer-Kuenen, W. Zhang, X. Zeng, Q. Li, M. Gaillard-Kelly, C. Wagner, and D. Tian, M. Aruva, W. Qin et al. Tian, A.
Chakrabarti, N. Amirkhanov et al. Aruva, K. Wang, W. Cai, K. Lyshchik, A. Fleischer, J. Huamani, D. Hallahan, M. Brissova, and J.
Cancer Imaging - Clinical Research - UCL EPSRC Centre for Doctoral Training In Medical Imaging
Levashova, M. Backer, C. Hamby, J. Pizzonia, J. Backer, and F. Chen, R. Tohme, A. Shahinian, J. Bading, and P. Chen, S. Hou et al. Cai, X. Zhang, Y. Wu, and X. Beer, M. Niemeyer, J. Carlsen et al. Komljenovic, M. Bretschi et al. View at Google Scholar P. Wu, X.
- Buckingham Palace Gardens: 25 (Grands détectives) (French Edition).
- Quantum Wells, Wires and Dots: Theoretical and Computational Physics of Semiconductor Nanostructures!
- Cecilia and the Scary Santa.
He, K. Zhang and X. Liu, W.
Improving prostate cancer diagnosis with MRI 'risk maps'
Hsieh, Y. Jiang et al. Bach-Gansmo, R. Danielsson, A. Saracco et al. Axelsson, T. Bach-Gansmo, J. Castell-Conesa, and B. Zarabi, M. Borgman, J. Zhuo, R. Gullapalli, and H. Munnink, W. Nagengast, A. Brouwers et al. S66—S73, Padmanabhan, J. Goggi, R. Bejot, and K. Sekar, A. Dhanabalan, and R. Yoo, C. Dence, T. Sharp, J. Katzenellenbogen, and M. Citri and Y. Capala and K. Scholl, P. Beuzeboc, and P. S81—S87, Chevallier, N. Rabillard, S. Wuilleme-Taumi, F. Harousseau, and H. Meric-Bernstam and M. Serrano-Olvera, A. Candelaria, and J. Tagliabue, A. Balsari, M. Campiglio, and S.
Zwick, C. Wallasch, and A. Ignatiadis, C. Desmedt, C. Sotiriou, E. De Azambuja, and M. Dijkers, J. Kosterink, A. Rademaker et al. Feldwisch, V. Tolmachev, C. Lendel et al. Tolmachev, J. Carlsson, S.