Diagnostic Imaging
April 2003


MRI may permit assessment of tumor growth processes

Noninvasive findings could more accurately stage cancer and monitor response to therapy

By: Arvind P. Pathak, Ph.D., Dmitri Artemov, Ph.D., Meiyappan Solaiyappan, and zaver m. bhujwalla, Ph.D.

Since its inception, MRI has been used routinely to detect and delineate tumors. Now, advanced MR techniques have been developed to measure the pathophysiology of angiogenesis. Researchers anticipate that MRI of this process will help to more accurately stage cancer and monitor its response to therapy.

Tumor angiogenesis, the process by which an avascular aggregate of tumor cells establishes a blood supply derived from the host stroma, is necessary for the establishment, growth, and metastasis of the tumor.1 Angiogenesis is a dynamic process, and a method to quantify neovascularization in situ without perturbing or excising the tumor could illuminate the temporal relationship between neovascularization and tumor growth.

A secondary set of events subsequent to cell proliferation appears to lead from hyperplasia to neoplasia, and the induction of angiogenesis is one such critical event, or "switch," in this cascade.2 During its prevascular stage, tumor growth to about 1 mm can occur in the absence of neovascularization, since all essential nutrients and waste products can diffuse across this distance. Once a tumor has reached this threshold stage, however, vascularization is essential for additional growth.3 The tumor acquires a blood supply by inducing neighboring preexisting microvasculature to sprout new capillaries that grow toward the tumor. To facilitate this angiogenesis, the tumor secretes chemicals collectively known as tumor angiogenesis factors (TAF) that diffuse into the surrounding tissue.

The initial reaction to this stimulus spurs endothelial cells in proximal vessels nearest to the chemical source to begin to alter their structure. Subsequently, enzymes degrade the basement membrane of the parent vessel, and capillary endothelial cells migrate toward the tumor, eventually developing into a new sprout in response to the TAF. This vascular phase is usually followed by rapid tumor growth, hemorrhage, and the potential for metastasis.

MRI, with its plethora of contrast mechanisms and superior dynamic functional range, has the potential to be a formidable tool in the noninvasive in vivo assessment of tumor angiogenesis. Researchers are learning how to use MRI to obtain an index of vascularization or microvessel density. They anticipate that these measures will be clinically advantageous for determining the effects of pre- or postoperative hormonal therapy on vascularization and the aggressiveness of the lesion, and for selecting and monitoring patients treated with antiangiogenic agents.4,5


Interest in tumor vasculature goes back almost a century,6 and it was recognized as early as 1945 that malignant cells provoke continuous vascular proliferation.7 Studies of tumor vascular morphology have identified a number of structural and functional differences between tumor and normal vasculature.8 Tumor-induced blood vessels are typically sinusoidal, fragile, and highly permeable, with a discontinuous basement membrane. Other characteristics of the tumor vasculature have also been noted:

- spatial heterogeneity and chaotic branching hierarchies;

- arteriovenous shunts;

- acutely as well as transiently collapsing vessels;

- poorly differentiated and leaky vessels lacking in smooth muscle cell lining; and

- inability to match the rapid proliferation of cancer cells, resulting in areas of hypoxia and necrosis.

These areas of hypoxic tissue are heterogeneously distributed within the tumor mass. A critical response to hypoxia, in normal and cancer cells, is the induction of the hypoxia-inducible transcription factor (HIF-1). One subunit of this factor, HIF-1a, is unique to HIF-1, and its expression is highly regulated by cellular oxygen concentration. Under hypoxic conditions, ubiquitination of HIF-1a is dramatically reduced and the activity of its transcriptional activation domain increases.9 HIF-1 can act as a transcriptional activator for vascular endothelial growth factor (VEGF) and several target genes that each contain hypoxia response elements (HRE), which include one or more HIF-1 binding sites. Thus HIF-1 mediates the adaptive response of cells to hypoxia.

VEGF is an angiogenic factor that appears to be most responsive to the abnormal physiological environments of hypoxia, extracellular acidosis, and substrate deprivation occurring in solid tumors. VEGF induces angiogenesis10 and is also a potent vascular permeability factor.11 Five isoforms of VEGF (or VEGF-A) bind to two tyrosine-kinase receptors, flt-1 and KDR/flk-1, which are expressed almost exclusively on endothelial cells. Hypoxia-mediated transcription of VEGF is believed to occur via the binding of HIF-1a to an HRE located in the VEGF promoter.12

VEGF inhibitors that make the tumor vasculature a potential therapeutic target for cancer therapy form the cornerstone of various antiangiogenesis clinical trials under way. The early successes of some of these therapies have highlighted the need to develop techniques to assess their efficacy in patients. MRI has the potential to provide a wealth of information relative to this goal by noninvasive means, with widespread clinical availability.

Bioassays have traditionally been used for studying angiogenesis in vivo. Three methods commonly used are the developing chick chorioallantoic membrane, the corneal pocket assay, and the transparent Millipore chamber model grown in the rabbit ear or hamster cheek pouch.13-15 Although these techniques have been used for pioneering studies in angiogenesis, their disadvantages include limitations on the number of samples that can be assayed, difficulty in quantifying and analyzing the results, invasiveness, and differences from the in vivo angiogenesis occurring in human tumors.


Tumor angiogenesis and vascularization can be visualized with MRI using either intrinsic or extrinsic contrast. Intrinsic, or endogenous, contrast arises from a chemical that is naturally present in the body and has sufficient magnetic susceptibility to influence the MR signal. An example is deoxyhemoglobin. Probing of the tumor vasculature using the intrinsic contrast produced by the deoxyhemoglobin in tumor microvessels is based on the blood oxygenation level-dependent (BOLD) contrast mechanism first proposed by Ogawa et al.16

The concentration of paramagnetic deoxyhemoglobin, the endogenous contrast agent in this case, is a primary determinant of the eventual image contrast. The presence of deoxyhemoglobin in a blood vessel causes a susceptibility difference between the vessel and its surrounding tissue. Such susceptibility differences induce microscopic magnetic field heterogeneities, or gradients, that cause dephasing of the MR proton signal, leading to a reduction in the value of T2*. In a T2*-weighted imaging experiment, the presence of deoxyhemoglobin in the blood vessels causes a darkening of the image in those voxels containing vessels. Since oxyhemoglobin is diamagnetic and does not produce the same dephasing, changes in oxygenation of the blood can be observed as signal changes in T2*-weighted images.

This technique has been used to detect changes in tumor oxygenation and hence vascularization following induction of angiogenesis by external angiogenic agents,17 as well as to obtain maps of the "functional" vasculature in genetically modified HIF-1 (+/+ and -/-) animal models.18 It should be kept in mind that BOLD contrast may not be solely related to deoxygenation but may be affected by other factors such as inflow, draining venules, hematocrit, oxygen saturation, blood flow, blood volume, and vessel orientation and geometry.19

One study demonstrated the feasibility of imaging blood flow in tumors using another endogenous contrast MR technique known as arterial spin labeling (ASL).20 In this approach, the arterial blood water itself served as a perfusion tracer. Although conducted in an animal model, the study strongly supported the use of ASL as a potential new method for monitoring the effects of therapies or agents designed to modulate tumor blood flow.

MRI using intrinsic contrast has the advantages of being completely noninvasive, not requiring the administration of any contrast agent or tracer pharmaceutical, and allowing the measurement of interest to be carried out as many times as desired, therefore providing dynamic data with high temporal resolution. These endogenous methods, however, cannot provide absolute quantitative measures of angiogenic parameters such as the tumor vascular volume or vascular permeability, which require exogenous contrast MRI.

Extrinsic, or exogenous, contrast arises from a chemical injected into the body to alter the local magnetic field in the tissue of interest, thereby enhancing the contrast in MR images. MR contrast agents are unique among diagnostic agents in that, unlike dyes or agents used with nuclear medicine or x-ray techniques, they are visualized not directly on the MR image but indirectly by virtue of the changes they induce in water relaxation behavior. The most commonly used contrast agents in MRI are the paramagnetic gadolinium chelates. The seven unpaired electrons of gadolinium produce a large magnetic moment that results in shortening of both the T1 (spin-lattice relaxation time) and T2 (spin-spin relaxation time) values of tissue water. Thus, on a T1-weighted MR scan, tissues that take up the paramagnetic agent are brightened (positive enhancement), while on a T2-weighted scan, the observed contrast is reversed (negative enhancement).

Using tracer kinetic or mass balance principles, the tissue concentration of gadolinium-based agents can be calculated from the MR image intensity. Several gadolinium complexes are under development or in use. They may be broadly classified as either low molecular weight ( almost equal to 0.57 kD), such as the Gd-DTPA compounds used clinically for contrast enhancement of various lesions including malignant tumors, or macromolecular ( almost equal to 65 kD), such as albumin-Gd-DTPA compounds that remain in the intravascular space for up to several hours.


Since low-molecular-weight contrast agents are the only class of paramagnetic agents approved for routine clinical use, several reports describe their application to image a variety of tumors including breast,21 brain,22 and uterine.23 Most T1-weighted methods involve analyses of relaxivity changes induced by the contrast agent to determine influx and outflux transfer constants, as well as the extracellular extravascular volume fractions on the basis of one of several theoretical models.24 A large range of clinically relevant tumor parameters can be inferred from these studies. T2- and T2*-weighted methods have also been employed in the study of tumor angiogenesis. These methods rely on the susceptibility effects of these contrast agents. More recently, in addition to the traditional tumor blood volume measurements by both T1-25 and T2-weighted26,27 techniques, measures of tumor vessel size28 and predictors of tumor grade29 have been added to the radiologist's repertoire.

The use of parameters may soon become more widespread in the clinical assessment of tumor angiogenesis with the advent of novel methods for the histological validation of these low-molecular-weight MR approaches,30 as well as techniques for correcting the effects of agent extravasation via leaky tumor vessels (Figure 1).27 A recent tumor xenograft study also demonstrated that the uptake kinetics of gadolinium may be used to successfully predict delivery of drug molecules (chemotherapeutic agents) to the interstitium of solid tumors (Figure 2).31


A drawback in the quantitation of tumor vascular parameters with low-molecular-weight contrast agents is their rapid extravasation from leaky tumor vessels. The availability of high-molecular-weight, or macromolecular, contrast agents (MMCA) such as albumin-Gd-DTPA complexes or synthetic compounds such as polylysine-Gd-DTPA and gadomer-17 provides an opportunity for the quantitative determination of tumor vascular volume and the permeability surface area (PS) product for molecules of comparable sizes.

The relatively slow leakage of the MMCA from the vasculature results in a long half-life and complete equilibration of plasma concentrations within the tumor. Assuming fast exchange of water between all the compartments in the tumor, the concentration of the MMCA within any given voxel is then proportional to the changes in relaxation rate (1/T1) before and after administration of the contrast agent. Relaxation rates can then be measured directly using either dynamic32 or steady-state T1-weighted methods.33

Voxel-wise maps constructed from the acquired data and fitted to an appropriate model provide spatial maps of tumor vascular volume and PS. Assuming a simple linear compartment model with negligible reflux of the contrast agent and constant blood concentrations of the agent for the duration of the MR experiment, the contrast uptake is a linear function of time. In this case, the slope of the concentration-time curve provides the parameter PS, and the y-intercept the vascular volume.34

Quantification of these parameters requires normalization to changes in the relaxation rate of the blood, which can be obtained separately from blood samples taken before administration of the contrast agent and again at the end of the MR experiment.

Several more complicated models of contrast uptake exist that are beyond the scope of this report. The accuracy of the tissue vascular volume measurement also depends on the rate of water exchange between the vascular and extracellular compartments.35

In addition to the aforementioned parameters, investigators have also employed the MMCA approach for detecting the efficacy of antiangiogenic treatments.36,37 More recently, MRI was used to detect vascular differences for metastatic versus nonmetastatic breast and prostrate cancer xenografts (Figure 3).38 The study used a noninvasive approach for assessing the role of the functional tumor vasculature in metastasis. These studies concluded that regions of high vascular volume were significantly less leaky compared with regions of low vascular volume.

Another conclusion of the study was that although an invasive approach was necessary, without adequate vascularization it was not sufficient for metastasis to occur.


The detection of tumor neovasculature by novel receptor-specific contrast agents has added a new dimension to MRI of angiogenesis. In one study, the contrast agent was composed of Gd-labeled polymerized liposomes conjugated with biotinylated antibodies targeted against aVb3 receptors on endothelial cells.39 On a standard T1-weighted image, the observed contrast was proportional to the density of receptors and thus to the density of the neovasculature.

Another recent study with breast cancer models demonstrated the feasibility of using MRI to detect the Her-2/neu receptor, an important cell surface receptor in breast cancer prognosis (Figure 4).40

MR has the potential ability to investigate the relationship that exists between tumor vascularization, physiology, and metabolism and to explain the impact that abnormal vasculature has on physiology, metabolism, and metastasis.

While it is useful to study these as separate characteristics, the ability to relate metabolism and vascularization within the same region of interest adds a new dimension to our understanding of how one impacts the other. Combined vascular and metabolic and (spectroscopic) MRI fulfills this need. The wealth of information afforded by MRI regarding angiogenic measurements has the potential to be of significant clinical importance in the noninvasive detection, diagnosis, prognosis, and treatment of cancer.

Dr. Pathak is a postdoctoral fellow of radiology, Dr. Artemov is an assistant professor of radiology, Mr. Solaiyappan is a research associate of radiology, and Dr. Bhujwalla is director of the In Vivo Cellular Molecular Imaging Program, all at Johns Hopkins University in Baltimore.


1. Baillie CT, Winslet MC, Bradley NJ. Tumor vasculature-a potential therapeutic target. Brit J Cancer 1995;72:257-267.

2. Folkman MJ, Watson K, Ingber D, Hanahan D. Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature 1989;339:58-61.

3. Folkman MJ. Angiogenesis: the unifying concept in cancer? J Natl Cancer Inst 1997;241:348-362.

4. Ezekowitz RAB, Mulliken JB, Folkman J. Interferon alfa-2a therapy for life-threatening hemangiomas of infancy. NEJM 1992;326:1456-1463.

5. O'Reilly MS, Boehm T, Shing Y, et al. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 1997;88:277-285.

6. Goldman E. The growth of malignant disease in man and the lower animals, with special reference to the vascular system. Proc R Soc Med 1997;1:1-13.

7. Algire GH, Chalkley HW, Legallais FY, Park HD. Vascular reactions of normal and malignant tissues in vivo. I. Vascular reactions of mice to wounds and to normal and neoplastic transplants. J Natl Cancer Inst 1945;6:73-85.

8. Konerding MA, van Ackern C, Fait E, et al. Morphological aspects of tumor angiogenesis and microcirculation. In: Molls M, Vaupel P, Brady LW, Heilmann HP, eds. Blood perfusion and microenvironment of human tumors: implications for clinical radiooncology (medical radiology). New York: Springer Verlag, 2000:5-17.

9. Semenza GL. Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr Opin Genet Dev 1998;8:588-594.

10. Gospodarowicz D, Abraham JA, Schilling J. Isolation and characterization of a vascular endothelial cell mitogen produced by pituitary-derived folliculo stellate cells. Proc Natl Acad Sci 1989;86:7311-7315.

11. Senger DR, Galli SJ, Dvorak AM, et al. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science 1983;219:983-985.

12. Levy AP, Levy NS, Wegner S, Goldberg MA. Transcriptional regulation of the rat endothelial growth factor gene by hypoxia. J Biol Chem 1995;270:13333-13340.

13. Ausprunk DH, Knighton DR, Folkman J. Vascularization of normal and neoplastic tissues grafted to the chick chorioallantois: role of host and preexisting graft blood vessels. Am J Pathol 1975;79:597-628.

14. Gimbone MA, Cotran RS, Leapman SB, Folkman J. Tumor growth and neovasculatization: an experimental model using the rabbit cornea. J Natl Cancer Inst 1974;52:413-427.

15. Greenblatt M, Shubik P. Tumor angiogenesis: transfilter diffusion studies in the hamster by the transparent chamber technique. J Natl Cancer Inst 1968;41:111-124.

16. Ogawa S. Oxygenation-sensitive contrast in MR image of rodent brain at high magnetic fields. Mag Reson Med 1990;14:68-78.

17. Abramovitch R, Frenkiel D, Neeman M. Analysis of subcutaneous angiogenesis by gradient echo magnetic resonance imaging. Mag Reson Med 1998;39:813-824.

18. Carmeliet P, Dor Y, Herbert J-M, et al. Role of HIF-1 in hypoxia-mediated apoptosis, cell proliferation and tumor angiogenesis. Nature 1998;394:485-490.

19. Weisskoff RM. Basic theoretical models of BOLD signal change. In: Moonen CTW, Bandettini PA, eds. Functional MRI. Heidelberg: Springer Verlag, 1999:115-125.

20. Silva AC, Kim S-G, Garwood M. Imaging blood flow in brain tumors using arterial spin labeling. Mag Reson Med 2000;44:169-173.

21. Furman-Haran E, Margalit R, Grobgeld D, Degani H. Dynamic contrast-enhanced magnetic resonance imaging reveals stress-induced angiogenesis in MCF7 human breast tumors. Proc Natl Acad Sci 1996;93:6247-6251.

22. Aronen HJ, Cohen MS, Belliveau JW, et al. Ultrafast imaging of brain tumors. Topics Mag Reson Imaging 1993;5:14-24.

23. Hawighorst H, Knapstein PG, Weikel W, et al. Angiogenesis of uterine cervical carcinoma: characterization by pharmacokinetic magnetic resonance parameters and histological microvessel density with correlation to lymphatic involvement. Cancer Res 1997;57:4777-4786.

24. Tofts PS. Modeling tracer kinetics in dynamic Gd-DTPA MR imaging. JMRI 1997;7:91-101.

25. Hacklander T, Reichenbach HR, Hofer M, Modder U. Measurement of cerebral blood volume via the relaxing effect low-dose gadopentate dimeglumine during bolus transit. AJNR 1996;17:821-830.

26. Aronen HJ, Gazit IE, Louis DN, et al. Cerebral blood volume maps of gliomas: comparison with tumor grade and histological findings. Radiology 1994;191:41-51.

27. Donahue KM, Krouwer HGJ, Rand SD, et al. Utility of simultaneously acquired gradient-echo and spin-echo cerebral blood volume and morphology maps in brain tumor patients. Mag Reson Med 2000;43:845-853.

28. Dennie J, Mandeville JB, Boxerman JL, et al. NMR imaging of changes in vascular morphology due to tumor angiogenesis. Mag Reson Med 1998;40:793-799.

29. Maeda M, Itoh S, Kimura H, et al. Tumor vascularity in the brain: evaluation with dynamic susceptibility-contrast MR imaging. Radiology 1993;189:233-238.

30. Pathak AP, Schmainda KM, Ward BD, et al. MR-derived cerebral blood volume maps: issues regarding histological validation and assessment of tumor angiogenesis. Mag Reson Med 2001;46:735-747.

31. Artemov D, Solaiyappan M, Bhujwalla ZM. Magnetic resonance pharmacoangiography to detect and predict chemotherapy delivery to solid tumors. Cancer Res 2001;61:3039-3044.

32. Schwarzbauer C, Syha J, Haase A. Quantification of regional cerebral blood volumes by rapid T1 mapping. Mag Reson Med 1993;29:709-712.

33. Brasch R, Pham C, Shames D, et al. Assessing tumor angiogenesis using macromolecular MR imaging contrast media. JMRI 1997;7:68-74.

34. Patlak CS, Blasberg RG, Fenstermacher JD. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. J Cereb Blood Flow Metab 1983;3:1-7.

35. Donahue KM, Weisskoff RM, Burstein D. Water diffusion and exchange as they influence contrast enhancement. J Magn Reson Imaging 1997;7:102-110.

36. Gossmann A, Helbich TH, Kuriyama N, et al. Dynamic contrast-enhanced magnetic resonance imaging as a surrogate marker of tumor response to anti-angiogenic therapy in a xenograft model of glioblastoma multiforme. JMRI 2002;15:233-240.

37. Bhujwalla ZM, Artemov D, Natarajan K, Kristjansen PEG. Anti-angiogenic agent TNP-470 significantly decreases permeable region. Glasgow, Scotland: 9th annual meeting, International Society for Magnetic Resonance in Medicine, 2001.

38. Bhujwalla ZM, Artemov D, Natarajan K, et al. Differences detected by MRI for metastatic versus nonmetastatic breast and prostate cancer xenografts. Neoplasia 2001;3:143-153.

39. Sipkins DA, Cheresh DA, Kazemi MR, et al. Detection of tumor angiogenesis in vivo by AB-targeted magnetic resonance imaging. Nat Med 1998;4:623-626.

40. Artemov D, Mori N, Ravi R, Bhujwalla ZM. MR molecular imaging of Her-2/neu receptor with Gd-based targeted contrast agent. Honolulu: 10th annual meeting, International Society for Magnetic Resonance in Medicine, 2002.