Mechanisms of = Tumor-related Brain Edema

http://www.medscape.com/ =

To Print: Click = your browser's=20 PRINT button.

NOTE: To view = the article=20 with Web enhancements, go to:
http://www.medscape.c= om/viewarticle/559000

Mechanisms of Tumor-related Brain Edema

Walter Stummer, M.D.

Neurosurg Focus. 2007;22(5):E8. = =A92007 American=20 Association of Neurological Surgeons

Posted 07/20/2007

Abstract and Introduction

Abstract

Cerebral edema contributes strongly to symptoms associated with = brain=20 tumors. Although the introduction of corticosteroids has greatly=20 simplified treatment of patients with newly diagnosed tumors, = these drugs=20 are associated with marked side effects during the long-term = treatment=20 that is often necessary in the recurrences. Therefore, a better=20 understanding of mechanisms related to the evolution and clearance = of=20 tumor-related edema with the aid of modern imaging and molecular=20 methodology is clearly necessary. Recently, researchers have = focused on=20 molecular mechanisms of edema development and have demonstrated=20 alternative routes=97such as the inhibition of vascular = endothelial growth=20 factor receptor inhibitors=97to be explored for treating edema. In = this=20 review the author focuses on established and current concepts = regarding=20 the pathophysiology of cerebral edema and its treatment.

Introduction

Edema associated with brain tumors plays a major role in = determining=20 symptoms caused by cerebral tumors.^[70] Not only does = edema=20 cause additional mass effect, often exceeding the mass induced by = the=20 tumor itself and resulting in increased intracranial pressure, it = also=20 leads to neurological disturbances by disrupting tissue = homeostasis and=20 reducing local blood flow.^[27,51,91] Near ly all focal = lesions,=20 including primary and metastatic tumors, abscesses, = encephalitides, and=20 radionecroses, pro- duce vasogenic edema. Although steroids have=20 facilitated the management of edema in patients with newly = diagnosed brain=20 tumors, long-term therapy of tumor-associated edema remains an = important=20 issue in the context of recurrent malignant gliomas or lesions = treated=20 using radiosurgery or radiotherapy rather than resection. = Nevertheless,=20 authors of only a few studies have specifically ad dressed = clinical=20 management issues of cerebral edema. Furthermore, viewing = tumor-associated=20 edema as a single entity is probably not justified. For example, = in=20 contrast to edema surrounding metastases (Fig. 1) or meningiomas,=20 peritumoral edema associated with high-grade gliomas (Fig. 2) is=20 characterized by extensive infiltration of tumor cells. Types of=20 peritumoral edema may therefore be distinguishable, depending on=20 histopathological and clinical prop erties. Adopting a more = differentiated=20 view would likely lead to improved strategies of treatment and = diagnosis=20 for brain tumors accompanied by peritumoral = edema.

Figure 1.

Figure 2.

In this review I summarize the established and new concepts = regarding=20 the pathophysiology of peritumoral edema based on morphological = and=20 molecular findings, as well as modern imaging.

Capillary Ultrastructure in Brain Tumors

Edema associated with brain tumors is considered to be = vasogenic and=20 thus pathophysiologically similar to edema due to brain injury or = cerebral=20 abscesses. The primary disturbance is at the level of the=20 microvasculature.^[50] In simplified terms, the tight = junctions=20 that form the BBB protect the brain's interstitial space from = plasma=20 extravasation under normal conditions, as there is no lymphatic = system=20 within the brain. In vasogenic edema, vascular permeability is = increased.=20 Under normal conditions a "sink effect" is provided by the = ventricles and=20 subarachnoid cerebrospinal fluid to allow steady circulation and=20 replenishment of the extracellular space. This sink effect is = overwhelmed=20 in vasogenic edema, resulting in extracellular fluid = accumulation.

Edema resulting from tumors must be distinguished from = cytotoxic edema,=20 for instance as a consequence of hypoxia of cytotoxic origin, = resulting=20 from cellular swell ing after breakdown of transmembraneous ion = gradients=20 due to energy depletion.^[40,50] These two types of = brain edema=20 were first differentiated by Klatzo^[38] in the = 1960s.

Metastatic and nonglial brain tumors produce angiogenic factors = that=20 promote capillary formations with marked ultrastructural = abnormalities. In=20 nonglial brain tumors, pinocytic vesicles are observed more = frequently,=20 and the basal laminae are more irregular.^[80] Glial = tumors also=20 lack a normal BBB. In a spheroid model of glioma in rats, the mass = of=20 tumor vessels display either abnormal tight junctions or=20 fenestrations.^[86] In human gliomas en dothelial cell = junctions=20 of capillaries appear either short or elongated, and the = endothelia=20 display hyperplasia, surface infolding of endothelial cells, = irregular=20 basal laminae, and large extravascular spaces in ultrathin and=20 freeze-fracture replicas.^[80] Proliferated endothelia = in=20 malignant gliomas are further characterized by fenestrations, = defective=20 tight junctions, increased numbers of pinocytic vesicles, and = incomplete=20 ensheathment of endothelial cells by the basal membrane. Although = vascular=20 permeability cannot be measured directly in tumor specimens from = humans,=20 it can be inferred that tumor vessels=97which are derived from = previously=20 existing vessels=97demonstrate structural abnormalities that = explain their=20 lost barrier functions, allowing leakage of plasma exudate into = the=20 surrounding brain.^{[28,44,45,55]}

On a molecular level, morphologically disrupted tight junctions = in=20 newly formed brain tumor capillaries are as sociated with a = paucity or=20 lack of proteins such as occludin, claudins, or the junctional = adhesion=20 molecule. These entities are all part of the molecular composition = of=20 tight junctions in the healthy = brain.^{[15,18,19,29,52,64]} These=20 transmembrane proteins bind intracellular proteins such as ZO-1 = and ZO-2.=20 Binding results in the coupling of tight junctions to the = cytoskeleton of=20 endothelial cells. It has been suggested that a decrease in = expression or=20 function of these tight junction proteins leads to opening of the = junction=20 and to the formation of edema. This hypothesis is sup ported by = findings=20 from several studies. For example, only low levels of claudin-1 = were found=20 to be expressed in mi crovessels from GBM, whereas high-grade = gliomas (WHO=20 Grades III and IV) did not express functional oc=20 cludin.^[42,64]

The Role of VEGF

In addition to structurally abnormal tumor vessels within the = tumor=20 itself, the vessels in neighboring tumor tissue may be affected by = infiltrating tumor cells and show clear ultrastructural changes, = such as=20 elongated junctional clefts (unfused regions) and an increase in = the=20 density of en dothelial vesicles, depending on the density of = infiltrating=20 cells. These changes were also observed when vessels were not = immediately=20 invested by tumor cells, suggesting the presence of cytokines, = which=20 spread into peritumoral tissue and disrupt normal vascular=20 morphology.^[85]

One cytokine that has been under close investigation is VEGF. = Paracrine=20 signal pathways involving VEGF may be involved in the generation = of=20 peritumoral edema. This cytokine was originally described as = vascular=20 permeability factor and is involved in angiogenesis and vascular=20 permeability.^[78] The VEGF binds to endothelial cells = via the=20 tyrosine kinase receptors flt-1 (VEGFR-1) and Flk-1/ KDR = (VEGFR-2). Both=20 are predominantly expressed on endothelial = cells.^[16,56]=20 Vascular endothelial growth factor appears to be involved in=20 tumorigenesis, neovascularization, and edema = production.^[48]=20 Apart from its mitogenic and chemotactic influence on endothelial = cells,=20 VEGF potently enhances permeability of endothelium and is ap = proximately=20 1000 times more potent than histamine.^[12,13] Similar = to=20 histamines and other mediators, VEGF exerts its effects on venules = and=20 small capillaries and acts immediately on endothelial cells, but = not on=20 smooth muscle cells, fibroblasts, or neutrophils, through = mobilization of=20 intracellular calcium.^[48] Vascular endothelial growth = factor=20 may impair the function of occludin by phosphorylation of this = protein,=20 with consecutive opening of tight junctions.^[63] = Possibly, VEGF=20 induces fenestration of the endo thelium and enhances capillary=20 permeability through this mechanism.^[72] On the other = hand,=20 upregulation of VEGF has been observed in brain tumors commonly = associated=20 with ede ma, such as GBMs, meningiomas, and=20 metastases.^[7,61,65,88] In patients with meningioma, = there is a=20 correlation be tween the presence or absence of peritumoral ede ma = and=20 levels of VEGF mRNA.^[22,37] Meningiomas with pronounced = edema=20 displayed higher levels of VEGF protein = staining.^[68]

Nevertheless, in the clinical situation clear correlations = between VEGF=20 and brain tumor grade or prognosis are missing.^[48] = Therefore,=20 the presence of edema does not automatically imply a high grade of = malignancy. For example, secretory meningiomas are benign = meningiomas (WHO=20 Grade I) and display extensive edema surrounding the=20 tumor.^[61,65,75] Meningiomas are not considered = severely=20 hypoxic, and other factors that upregulate VEGF may be involved, = such as=20 platelet-derived growth factor, epidermal growth factor, and=20 estrogens.^[48,53,83] Ede ma in meningiomas has also = been=20 associated with morphological factors, including tumor size and=20 histological subtype, whereby transitional and meningotheliomatous = types=20 tend to produce more edema.^[21] Furthermore, invasion = of=20 adjacent brain by meningioma has been related to the presence of=20 edema.^[8,32,49] The invading meningioma phenotype = disrupts the=20 bordering arachnoid, possibly providing a route for the spread of = edema=20 mediators, including VEGF. These mediators may cause not only the=20 propagation of edema but also the proliferation of new blood = vessels, thus=20 explaining the high correlation be tween pial blood supply and = peritumoral=20 brain edema.

In contrast, VEGF expression in human gliomas is related to = their=20 degree of malignancy. Low-grade gliomas express low levels of = VEGF,=20 whereas progression of low-grade gliomas into malignant gliomas is = associated with an up to 50-fold increase in VEGF mRNA=20 levels.^[66] In creased VEGF expression has been = correlated with=20 vascular permeability in situ.^[46] The expression of = VEGF is=20 largely restricted to perinecrotic cells, suggesting that hy poxia = regulates its expression in high-grade gliomas.^[82] = Conversly,=20 pilocytic astrocytomas (WHO Grade I) have an excellent prognosis = while=20 displaying high levels of VEGF mRNA.^[41] Regulation of = VEGF in=20 these tumors may be the consequence of factors other than hypoxia, = such as=20 a loss of function of the p53 tumor suppressor=20 gene.^[58]

Other Mechanisms Under Investigation

Apart from VEGF, other factors implicated in edema formation = related to=20 brain tumors include arachidonic acid metabolites and NO. Elevated = levels=20 of leukotriene C4 (produced via the lipoxygenase pathway) have = been found=20 in GBM and in adjacent edematous brain tissue in correlation with = the=20 amount of peritumoral edema.^[9,59] In a rat glioma = model,=20 microglia, which infiltrates brain tumors, was demonstrated to be = a major=20 source of prostaglandin E2 production through the COX-2=20 pathway.^[3] The COX-2- and lipoxygenase-derived = biologically=20 active lipid mediators are thought to promote tumorigenesis and=20 peritumoral brain edema.^[59]

Nitric oxide has been identified as a specific mediator of = vasodilation=20 and tumor blood flow in primary brain tu mors and has also been=20 investigated in the context of brain tumor-associated edema. The = concept=20 involves induction of NOS isoenzymes as a result of tumor hypoxia. = In=20 ducible NOS II is expressed in tumor-infiltrating micro glia and=20 macrophages. However, a correlation between the quantity and = expression of=20 NOS has not been demonstrated in primary brain tumors, whereas a = specific=20 edema-modulating role of NO in cerebral metastasis has been con=20 sidered.^[10] In this context, it appears that = VEGF-induced=20 edema formation may occur via the synthesis and release of=20 NO.^[54] Other vasogenic substances that may contribute = to the=20 pathophysiology of tumor-related edema are serotonin, = thromboxanes, and=20 platelet-activating factor.^[20,30]

Macrophages may also be involved in edema accumulation because = the=20 degree of macrophage infiltration seen on immunoperoxidase = staining=20 correlates with the extent of peritumoral edema.^[81] = This has=20 been taken to suggest that secretory products of macrophages might = contribute to edema formation associated with brain tumors.

An additional hypothesis with possible implications for the = formation=20 and regulation of brain edema has recently emerged after the = discovery of=20 the aquaporin protein family. Aquaporins are small, hydrophobic, = integral=20 membrane proteins that are expressed in all living organisms and = play=20 critical roles in controlling the water flow into and out of = cells. In the=20 brain, aquaporin-4 is expressed in endothelial astrocytic foot = processes.=20 Aquaporin-4 is highly upregulated in high-grade = gliomas.^[60,74]=20 As yet, however, it is undetermined whether this upregulation = results in=20 increased edema formation or in enhanced clearance of edema. = Nevertheless,=20 modulation of the expression and/or function of aquaporins may = provide=20 novel therapeutic options for reducing brain tumor=96associated=20 edema.

Bulk Flow of Edema

One of the first pathophysiological aspects of tumor-associated = edema=20 to be studied extensively was bulk flow. Interest in bulk flow has = recently been revived due to the development of = convection-enhanced=20 delivery as a treatment modality for malignant gliomas, as well as = to ad=20 vances in neuroradiology, such as DT imaging. His torically, the=20 introduction of CT in the 1970s advanced the understanding of = edema=20 propagation and resolution in patients. Around that time it was=20 established that tumor-associated edema propagates by bulk flow = rather=20 than by simple diffusion. Due to bulk flow, small pressure = gradients, such=20 as the gradient between gray and white matter, are sufficient to = prevent=20 or redirect the spreading of edema. White matter has a lower = resistance=20 than the more densely packed gray matter; therefore, edema tends = to=20 propagate in fingerlike projections throughout the white=20 matter.^[69] This phenomenon has been confirmed with=20 mathematical models based on CT measurements.^[1] = Computed=20 tomography has also been used to measure the speed of = extravasation of=20 intravenous contrast material.^[34] Based on a spherical = model=20 of edema propagation from the lesion, the formation rate of edema = fluid=20 was calculated as 0.5 to 3.2 ml/hour and the speed of edema fluid=20 spreading to 1.9 mm/hour. In a patient with a metastasis, fluid = was=20 produced at a rate of 3.9 ml/hour, which led to an edematous = volume in the=20 brain of 87.3 ml.^[70]

In these key analyses it was recognized that the volume of = edema=20 depends not only on the rate of edema production but also on the = rate of=20 resorption of extravasated fluid. Clearance of extravasated fluid = was=20 considered to be due to the hydrostatic pressure responsible for = the bulk=20 flow of edema fluid pushing the fluid away from the lesion until = it is=20 cleared primarily in the ventricular^[71] and, to a = lesser=20 extent, the subarachnoid cerebrospinal fluid.^[95] Other = clearance mechanisms include absorption through the subarachnoid = space,=20 which has been demonstrated to be independent of intracranial=20 pressure^[95] or=97to a limited extent=97the resorption = of=20 extravasated proteins by astrocytic cells. This mechanism = decreases the=20 extracellular on cotic pressure and thus limits propagation, = contributing=20 to the resolution of edema.^[24] Finally, there is a = limited=20 amount of absorption by local capillaries in the edematous tissue, = which=20 can be quantified (authors of one study found this rate to be = 0.0086=20 ml/hr/cm^[3]).^[24] Work involving edema = formation and=20 resolution before and after dexa methasone treatment has shown = that the=20 drug acts by de creasing edema formation rather than by increasing = its=20 resolution.^[33]

The more recent emergence of MR imaging as a diagnostic tool = has=20 corroborated the earlier CT findings and has given access to = additional=20 information regarding the pathophysiology of tumor-related edema. = For=20 instance, by measuring changes in relaxation time (T1), the = extracellular=20 distribution volume (V_d) and capillary permeability = expressed=20 as the unidirectional transport rate constant (K_i) can = be=20 calculated for various tumor types. Meningiomas have a higher=20 K_i than gliomas or metastases. After treatment with=20 dexamethasone, the K_i decreases by 52% in gliomas and=20 metastases, but only by 4% in meningiomas.^[2] Magnetic=20 resonance imaging has also been used to measure cerebral blood = flow,=20 cerebral blood volume, and brain=96tumor barrier permeability. = Changes in=20 perfusion have been studied using dynamic perfusion-weighted MR = imaging.=20 Relative regional cerebral blood volume and flow were almost 50% = lower in=20 peritumoral brain edema than in the contralateral white matter in = patients=20 with meningiomas.^[91] Furthermore, the positive effects = of=20 dexamethasone administration on cerebral blood flow and BBB = permeability=20 have been quantified by MR imaging.^[4,62]

More novel assessments of peritumoral edema involve DT imaging = for=20 making visible and analyzing apparent water diffusivity in tissues = in=20 vivo.^[5,39,57] Using this technique, investigators have = found=20 greater apparent water diffusivity in high-grade gliomas, whereas = the=20 anisotropy (the property of being directionally dependent) in = these le=20 sions was comparable to that of edema in other tumors (low-grade = gliomas,=20 metastases, and meningiomas). This observation was taken to imply = that=20 there are two distinct types of peritumoral edema: edema = associated with=20 high-grade gliomas and edema associated with low-grade gliomas or = nonglial=20 tumors, despite the lack of a signal intensity difference between = these=20 two edema groups on T2-weighted MR imaging. Water movement in = areas of=20 edema, predominantly in the extracellular spaces, was less = restricted in=20 high-grade gliomas, a phenomenon that likely reflected the = destruction of=20 the extracellular matrix ultrastructure by malignant cell = infiltration=20 and, consequently, greater water diffusion. On the basis of these=20 observations, it has been argued that DT imaging could be used as = a=20 clinical tool for differentiating high-grade gliomas and for = evaluating=20 the extent of cellular infiltration.

Correlative studies on the histopathological characteristics of = malignant gliomas and imaging findings have demonstrated a high = degree of=20 concordance between the T2 signal abnormality on MR images as an = indicator=20 of edema and tumor cell infiltration.^[93] Fractional = anisotropy=20 seen on DT images correlated with the density of tumor cell in=20 filtration.^[84] Therefore, a strong association between = edema=20 propagation and malignant cell infiltration has been shown in = these=20 studies.

This understanding of cell migration and fluid flow dynamics = within=20 tumor and brain interstitial space has gained attention in the = context of=20 convection-enhanced delivery, which has been developed as a novel=20 instrument for transporting drugs to infiltrating malignant glioma = cells.^[25] Therefore, methods of assessment and = visualization=20 of fluid movements within the brain, such as DT imaging, will be = of future=20 interest, not only for optimizing edema treatment, but also for = optimizing=20 brain tumor therapy.^[76]

Management of Brain Tumor Edema

The mainstay of therapy of brain tumor=96related edema is = corticosteroids=20 for both nonglial and glial tumors. These drugs have been used = since the=20 1960s^[36,73] and have led to a remarkable decline in=20 perioperative mortality and morbidity rates. The preferred = compound is=20 dexamethasone due to its low mineralocorticoid effects in = comparison with=20 other corticosteroids. With long-term use, however, the positive = action is=20 counterweighted by deleterious side effects such as = immunosuppression,=20 weight gain, gastrointestinal problems, osteoporosis, myopathy, = and an=20 enhanced risk of deep vein thrombosis and pulmonary em- bolism. = The=20 mechanisms of action of corticosteroids are still unclear. = Corticosteroids=20 lead to a decrease in the rate of edema formation without = effecting=20 clearance,^[33] which appears to occur within 1 hour = after=20 administration, as demonstrated by reductions in capillary=20 permeability.^[79] Al though data have been presented = showing=20 that dexa methasone acts by reducing the expression of=20 VEGF,^[26,47] this agent may act by interfering with = VEGF action=20 on the target endothelial cell.^[55] In cultured = endothelial=20 cells, VEGF has been shown to increase intracellular = Ca²⁺,=20 which has been associated with cytoskeletal rearrangement. = Dexamethasone=20 appears to block this influx.^[14] Furthermore, in = peripheral=20 blood vessels, the increase in vascular permeability seen after=20 intradermal injection of VEGF is inhibited by systemic = administration of=20 dexamethasone, acting through the glucocorticoid=20 receptor.^[11,26]

In Western Europe, neurooncologists try to treat brain = tumor=96induced=20 edema and to overcome corticosteroid dependency as well as=20 corticosteroid-related side effects by administering boswellic = acids=20 (H15). These are phytotherapeutic agents that are believed to = inhibit=20 edema formation and even tumor growth in patients with malignant=20 gliomas.^[35,87]

A number of additional experimental approaches are being = pursued at the=20 moment; some are still in the preclinical stage, and some are in = early=20 clinical studies. Thrombin has been investigated as a possible = target of=20 antiedema strategies. An enzyme involved in the coagulation = cascade,=20 thrombin has also been implicated in brain edema formation, = angiogenesis,=20 and cell proliferation. Argatroban, a thrombin antagonist, was = shown to=20 reduce edema, tumor growth, and tumor-related neurological = deficits in rat=20 glioma models.^[23,31]

Attention has also been directed toward corticotropin- = releasing=20 factor, a polypeptide that regulates adrenocorticotropin hormone = release=20 from the pituitary gland, which in turn regulates hydrocortisone = secretion=20 from the adrenal gland. Corticotropin-releasing factor has also = been=20 studied in conjunction with intracerebral gliomas and has been = found to=20 reduce BBB permeability in a tumor model^[90] and to = improve=20 neurological function in patients suffering brain=20 metastasis.^[92]

Due to its implication in the generation of brain = tumor=96related edema,=20 selective inhibition of VEGF has been explored, for instance by = the=20 antiangiogenic drug SU5416 (semaxanib), a small molecule that = selectively=20 inhibits tyrosine kinase activity of the VEGFR Flk-1/ KDR. This = drug has=20 been associated with prolonged survival in rats with intracerebral = gliosarcoma and increased necrosis while reducing=20 vascularity.^[89] Similar observations were made in nude = mice=20 with human GBM xenografts.^[77] Although promising = activity in=20 neoplastic disease outside the brain has been demonstrated in=20 patients,^[17,43,96] the value of semaxanib in brain = edema=20 treatment is unclear. On the other hand, the pan-VEGFR inhibitor = AZD2171=20 has to date shown efficacy in normalizing vasculature and = alleviating=20 edema in patients with GBM,^[6] giving proof of the = principle of=20 this concept and endorsing further studies.

Another class of compounds has recently received some = attention, the=20 inhibitors of COX2. In one study, the COX-2 inhibitor SC-236 has=20 demonstrated a prolongation of survival similar to dexamethasone = when=20 administered in rats with intracerebral = gliosarcomas.^[67]=20 Similarily, the diffusion of contrast medium into adjacent brain = was=20 attenuated in rats with intracerebral C6-gliomas after = pretreatment with=20 rofecoxib, another selective COX-2 in hibitor. Rofecoxib was as = effective=20 as dexamethasone in these experiments. A mode of action for COX-2=20 inhibitors might be the reduction of transcription factor Sp1's = DNA=20 binding and transactivating activity, resulting in less production = of=20 VEGF.^[94]

Future Directions

To date, VEGFR antagonists, COX-2 inhibitors, boswellic acid, = and other=20 experimental therapeutic agents have not been shown to be superior = to=20 corticosteroids in the management of brain tumor=96related edema;=20 furthermore, careful assessments are required. A better = understanding of=20 different types of tumor edema related to different pathological = entities=20 and their molecular mechanisms may open the way to targeted = therapies. Im=20 proved understanding of the molecular mechanisms controlling BBB=20 permeability will allow manipulation of this permeability for = improving=20 therapy for malignant brain tumors with drugs that, under = physiological=20 circumstances, cannot cross this important barrier.

References

Aaslid R, Groger U, Patlak CS, Fenstermacher JD, Huber = P,=20 Reulen HJ: Fluid flow rates in human peritumoral oedema. = Acta=20 Neurochir Suppl (Wien) 51:152=96154, 1990
Andersen C: The effect of glucocorticoids in the = normal=20 cerebral hemisphere of brain tumor patients. Acta Neurol = Scand=20 98:433=96438, 1998
Badie B, Schartner JM, Hagar AR, Prabakaran S, Peebles = TR,=20 Bartley B, et al: Microglia cyclooxygenase-2 activity in=20 experimental gliomas: possible role in cerebral edema = formation. J=20 Clin Cancer Res 9:872=96877, 2003
Bastin ME, Carpenter TK, Armitage PA, Sinha S, Wardlaw = JM,=20 Whittle IR: Effects of dexamethasone on cerebral perfusion = and=20 water diffusion in patients with high-grade glioma. AJNR = Am J=20 Neuroradiol 27:402=96408, 2006
Bastin ME, Sinha S, Whittle IR, Wardlaw JM: = Measurements of=20 water diffusion and T1 values in peritumoral oedematous = brain.=20 Neuroreport 13:1335=961340, 2002
Batchelor TT, Sorensen AG, di Tomaso E, Zhang WT, Duda = DG,=20 Cohen KS, et al: AZD2171, a pan-VEGF receptor tyrosine = kinase=20 inhibitor, normalizes tumor vasculature and alleviates = edema in=20 glioblastoma patients. Cancer Cell 11:83=9695, 2007
Berkman RA, Merrill MJ, Reinhold WC, Monacci WT, = Saxena A,=20 Clark WC, et al: Expression of the vascular permeability=20 factor/vascular endothelial growth factor gene in central = nervous=20 system neoplasms. J Clin Invest 91:153=96159, 1993
Bitzer M, Wockel L, Morgalla M, Keller C, Friese S, = Heiss E,=20 et al: Peritumoral brain oedema in intracranial = meningiomas:=20 influence of tumor size, location and histology. Acta Neu = ro chir=20 (Wien) 139:1136=961142, 1997
Black KL, Hoff JT, McGillicuddy JE, Gebarski SS: = Increased=20 leukotriene C4 and vasogenic edema surrounding brain = tumors in=20 humans. Ann Neurol 19:592=96595, 1986
Broholm H, Rubin I, Kruse A, Braendstrup O, Schmidt K, = Skri=20 ver EB, et al: Nitric oxide synthase expression and = enzymatic=20 activity in human brain tumors. Clin Neuropathol 22: = 273=96281, 2003
Bruce JN, Criscuolo GR, Merrill MJ, Moquin RR, = Blacklock JB,=20 Oldfield EH: Vascular permeability induced by protein = product of=20 malignant brain tumors: inhibition by dexamethasone. J = Neurosurg=20 67:880=96884, 1987
Connolly DT: Vascular permeability factor: a unique = regulator=20 of blood vessel function. J Cell Biochem 47:219=96223, = 1991
Criscuolo GR: The genesis of peritumoral vasogenic = brain edema=20 and tumor cysts, a hypothetical role for tumor-derived = vascular=20 permeability factor. Yale J Biol Med 66:277=96314, 1993
Criscuolo GR, Lelkes PI, Rotrosen D, Oldfield EH: = Cytosolic=20 calcium changes in endothelial cells induced by a protein = product=20 of human gliomas containing vascular permeability factor = activity.=20 J Neurosurg 71:884=96891, 1989
Davies DC: Blood-brain barrier breakdown in septic en=20 cephalopathy and brain tumors. J Anat 200:639=96646, 2002
de Vries C, Escobedo JA, Ueno H, Houck K, Ferrara N, = Williams=20 LT: The fms-like tyrosine kinase, a receptor for vascular=20 endothelial growth factor. Science 255:989=96991, 1992
Fiedler W, Mesters R, Tinnefeld H, Loges S, Staib P, = Duhrsen=20 U, et al: A phase 2 clinical study of SU5416 in patients = with=20 refractory acute myeloid leukemia. Blood 102:2763=962767, = 2003
Furuse M, Fujita K, Hiiragi T, Fujimoto K, Tsukita S:=20 Claudin-1 and -2: novel integral membrane proteins = localizing at=20 tight junctions with no sequence similarity to occludin. J = Cell=20 Biol 141:1539=961550, 1998
Furuse M, Hirase T, Itoh M, Nagafuchi A, Yonemura S, = Tsukita=20 S, et al: Occludin: a novel integral membrane protein = localizing=20 at tight junctions. J Cell Biol 123:1777=961788, 1993
Gaetani P, Butti G, Chiabrando C, Danova M, Castelli = MG,=20 Riccardi A, et al: A study on the biological behavior of = human=20 brain tumors. Part I. Arachidonic acid metabolism and DNA = content.=20 J Neurooncol 10:233=96240, 1991
Go KG, Wilmink JT, Molenaar WM: Peritumoral brain = edema=20 associated with meningiomas. Neurosurgery 23:175=96179, = 1988
Goldman CK, Bharara S, Palmer CA, Vitek J, Tsai JC, = Weiss HL,=20 et al: Brain edema in meningiomas is associated with in = creased=20 vascular endothelial growth factor expression. = Neurosurgery=20 40:1269=961277, 1997
Gorman AM, Hirt UA, Orrenius S, Ceccatelli S: Dexa = meth asone=20 pre-treatment interferes with apoptotic death in glioma = cells.=20 Neuroscience 96:417=96425, 2000
Groger U, Huber P, Reulen HJ: Formation and resolution = of=20 human peritumoral brain edema. Acta Neurochir Suppl 60: = 373=96374,=20 1994
Hall WA, Sherr GT: Convection-enhanced delivery: = targeted=20 toxin treatment of malignant glioma. Neurosurg Focus = 20(4): E10,=20 2006
Heiss JD, Papavassiliou E, Merrill MJ, Nieman L, = Knightly JJ,=20 Walbridge S, et al: Mechanism of dexamethasone suppression = of=20 brain tumor-associated vascular permeability in rats. = Involvement=20 of the glucocorticoid receptor and vascular permeability = factor. J=20 Clin Invest 98:1400=961408, 1996
Hino A, Imahori Y, Tenjin H, Mizukawa N, Ueda S, = Hirakawa K,=20 et al: Metabolic and hemodynamic aspects of peritumoral=20 low-density areas in human brain tumor. Neurosurgery 26: = 615=96621,=20 1990
Hirano A, Kawanami T, Llena JF: Electron microscopy of = the=20 blood-brain barrier in disease. Microsc Res Tech = 27:543=96556, 1994
Hirase T, Staddon JM, Saitou M, Ando-Akatsuka Y, Itoh = M,=20 Furuse M, et al: Occludin as a possible determinant of = tight=20 junction permeability in endothelial cells. J Cell Sci = 110:=20 1603=961613, 1997
Hirashima Y, Hayashi N, Fukuda O, Ito H, Endo S, = Takaku A:=20 Platelet-activating factor and edema surrounding = meningiomas. J=20 Neurosurg 88:304=96307, 1998
Hua Y, Tang LL, Fewel ME, Keep RF, Schallert T, = Muraszko KM,=20 et al: Systemic use of argatroban reduces tumor mass, = attenuates=20 neurological deficits and prolongs survival time in rat = glioma=20 models. Acta Neurochir Suppl 95:403=96406, 2005
Ide M, Jimbo M, Kubo O, Yamamoto M, Takeyama E, = Imanaga H:=20 Peritumoral brain edema and cortical damage by meningioma. = Acta=20 Neurochir Suppl 60:369=96372, 1994
Ito U, Reulen HJ, Tomita H, Ikeda J, Saito J, Maehara = T: A com=20 puted tomography study on formation, propagation, and = resolution=20 of edema fluid in metastatic brain tumors. Adv Neu rol = 52:459=96468,=20 1990
Ito U, Reulen HJ, Tomita H, Ikeda J, Saito J, Maehara = T: For=20 mation and propagation of brain oedema fluid around human = brain=20 metastases: a CT study. Acta Neurochir (Wien) 90: 35=9641, = 1988
Janssen G, Bode U, Breu H, Dohrn B, Engelbrecht V, = Gobel U:=20 Boswellic acids in the palliative therapy of children with = progressive or relapsed brain tumors. Klin PE4diatr = 212:189=96195,=20 2000
Jelsma R, Bucy PC: The treatment of glioblastoma = multiforme of=20 the brain. J Neurosurg 27:388=96400, 1967
Kalkanis SN, Carroll RS, Zhang J, Zamani AA, Black PM: = Correlation of vascular endothelial growth factor = messenger RNA=20 expression with peritumoral vasogenic cerebral edema in=20 meningiomas. J Neurosurg 85:1095=961101, 1996
Klatzo I: Presidential address. Neuropathological = aspects of=20 brain edema. J Neuropathol Exp Neurol 26:1=9614, 1967
Kono K, Inoue Y, Nakayama K, Shakudo M, Morino M, = Ohata K, et=20 al: The role of diffusion-weighted imaging in patients = with brain=20 tumors. AJNR Am J Neuroradiol 22:1081=961088, 2001
Kucinsky T, Vaterlein O, Glauche V, Fiehler J, Klotz = E, Eckert=20 B, et al: Correlation of apparent diffusion coefficient = and=20 computed tomography density in acute ischemic stroke. = Stroke 33:=20 1786=961791, 2002
Leung SY, Chan AS, Wong MP, Yuen ST, Cheung N, Chung = LP:=20 Expression of vascular endothelial growth factor and its = receptors=20 in pilocytic astrocytoma. Am J Surg Pathol 21: 941=96950, = 1997
Liebner S, Fischmann A, Rascher G, Duffner F, Grote = EH,=20 Kalbacher H, et al: Claudin-1 and claudin-5 expression and = tight=20 junction morphology are altered in blood vessels of human=20 glioblastoma multiforme. Acta Neuropathol 100:323=96331, = 2000
Loges S, Tinnefeld H, Metzner A, Jucker M, Butzal M, = Bru=20 weleit M, et al: Downregulation of VEGF-A, STAT5 and AKT = in acute=20 myeloid leukemia blasts of patients treated with SU5416. = Leuk=20 Lymphoma 47:2601=962609, 2006
Long DM: Capillary ultrastructure and the blood-brain = barrier=20 in human malignant brain tumors. J Neurosurg 32:127=96144, = 1970
Long DM: Capillary ultrastructure in human metastatic = brain=20 tumors. J Neurosurg 51:53=9658, 1979
Machein MR, Kullmer J, Fiebich BL, Plate KH, Warnke = PC:=20 Vascular endothelial growth factor expression, vascular = volume,=20 and capillary permeability in human brain tumors. Neu = rosurgery=20 44:732=96741, 1999
Machein MR, Kullmer J, Ronicke V, Machein U, Krieg M, = Damert=20 A, et al: Differential downregulation of vascular = endothelial=20 growth factor by dexamethasone in normoxic and hypoxic rat = glioma=20 cells. Neuropathol Appl Neurobiol 25: 104=96112, 1999
Machein MR, Plate KH: VEGF in brain tumors. J = Neurooncol=20 50:109=96120, 2000
Mantle RE, Lach B, Delgado MR, Baeesa S, Belanger G:=20 Predicting the probability of meningioma recurrence based = on the=20 quantity of peritumoral brain edema on computerized to = mography=20 scanning. J Neurosurg 91:375=96383, 1999
Marmarou A, Portella G, Barzo P, Signoretti S, = Fatouros P,=20 Beaumont A, et al: Distinguishing between cellular and = vasogenic=20 edema in head injured patients with focal lesions using = magnetic=20 resonance imaging. Acta Neurochir Suppl 76: 349=96351, = 2000
Marmarou A, Takagi H, Shulman K: Biomechanics of brain = edema=20 and effects on local cerebral blood flow. Adv Neurol = 28:345=96358,=20 1980
Martin-Padura I, Lostaglio S, Schneemann M, Williams = L, Romano=20 M, Fruscella P, et al: Junctional adhesion molecule, a = novel=20 member of the immunoglobulin superfamily that distributes = at=20 intercellular junctions and modulates monocyte = transmigration. J=20 Cell Biol 142:117=96127, 1998
Maxwell M, Galanopoulos T, Hedley-Whyte ET, Black PM,=20 Antoniades HN: Human meningiomas co-express = platelet-derived=20 growth factor (PDGF) and PDGF-receptor genes and their = protein=20 products. Int J Cancer 46:16=9621, 1990
Mayhan WG: VEGF increases permeability of the = blood-brain=20 barrier via a nitric oxide synthase/cGMP-dependent = pathway. Am J=20 Physiol 276:C1148=96C1153, 1999
Merrill MJ, Oldfield EH: A reassessment of vascular=20 endothelial growth factor in central nervous system = pathology. J=20 Neurosurg 103:853=96868, 2005
Millauer B, Wizigmann-Voos S, Schnurch H, Martinez R, = Moller=20 NP, Risau W, et al: High affinity VEGF binding and = developmental=20 expression suggest Flk-1 as a major regulator of = vasculogenesis=20 and angiogenesis. Cell 72:835=96846, 1993
Morita K, Matsuzawa H, Fujii Y, Tanaka R, Kwee IL, = Nakada T:=20 Diffusion tensor analysis of peritumoral edema using = lambda chart=20 analysis indicative of the heterogeneity of the = microstructure=20 within edema. J Neurosurg 102:336=96341, 2005
Mukhopadhyay D, Tsiokas L, Sukhatme VP: Wild-type p53 = and=20 v-Src exert opposing influences on human vascular = endothelial=20 growth factor gene expression. Cancer Res 55:6161=966165, = 1995
Nathoo N, Barnett GH, Golubic M: The eicosanoid = cascade:=20 possible role in gliomas and meningiomas. J Clin Pathol = 57:6=9613,=20 2004
Nielsen S, Nagelhus EA, Amiry-Moghaddam M, Bourque C, = Agre P,=20 Ottersen OP: Specialized membrane domains for water = transport in=20 glial cells: high-resolution immunogold cytochemistry of=20 aquaporin-4 in rat brain. J Neurosci 17:171=96180, 1997
Nishikawa R, Cheng SY, Nagashima R, Huang HJ, Cavenee = WK,=20 Matsutani M: Expression of vascular endothelial growth = factor in=20 human brain tumors. Acta Neuropathol 96:453=96462, 1998
Ostergaard L, Hochberg FH, Rabinov JD, Sorensen AG, = Lev M, Kim=20 L, et al: Early changes measured by magnetic resonance = imaging in=20 cerebral blood flow, blood volume, and blood-brain barrier = permeability following dexamethasone treatment in patients = with=20 brain tumors. J Neurosurg 90:300=96305, 1999
Papadopoulos MC, Saadoun S, Davies DC, Bell BA: = Emerging=20 molecular mechanisms of brain tumor oedema. Br J Neurosurg = 15:101=96108, 2001
Papadopoulos MC, Saadoun S, Woodrow CJ, Davies DC,=20 Costa-Martins P, Moss RF, et al: Occludin expression in=20 microvessels of neoplastic and non-neoplastic human brain. = Neuropathol Appl Neurobiol 27:384=96395, 2001
Pietsch T, Valter MM, Wolf HK, von Deimling A, Huang = HJ,=20 Cavenee WK, et al: Expression and distribution of vascular = endothelial growth factor protein in human brain tumors. = Acta=20 Neuropathol 93:109=96117, 1997
Plate KH, Breier G, Weich HA, Risau W: Vascular = endothelial=20 growth factor is a potential tumor angiogenesis factor in = human=20 gliomas in vivo. Nature 359:845=96848, 1992
Portnow J, Suleman S, Grossman SA, Eller S, Carson K: = A=20 cyclooxygenase-2 (COX-2) inhibitor compared with dexa = methasone in=20 a survival study of rats with intracerebral 9L = gliosarcomas.=20 Neuro-oncol 4:22=9625, 2002
Provias J, Claffey K, delAguila L, Lau N, Feldkamp M, = Guha A:=20 Meningiomas: role of vascular endothelial growth factor/ = vascular=20 permeability factor in angiogenesis and peritumoral edema. = Neurosurgery 40:1016=961026, 1997
Reulen HJ, Graham R, Spatz M, Klatzo I: Role of = pressure=20 gradients and bulk flow in dynamics of vasogenic brain = edema. J=20 Neurosurg 46:24=9635, 1977
Reulen HJ, Huber P, Ito U, Groger U: Peritumoral brain = edema.=20 A keynote address. Adv Neurol 52:307=96315, 1990
Reulen HJ, Tsuyumu M, Tack A, Fenske AR, Prioleau GR:=20 Clearance of edema fluid into cerebrospinal fluid. A = mechanism for=20 resolution of vasogenic brain edema. J Neurosurg = 48:754=96764, 1978
Roberts WG, Palade GE: Neovasculature induced by = vascular=20 endothelial growth factor is fenestrated. Cancer Res = 57:765=96772,=20 1997
Ruderman NB, Hall TC: Use of glucocorticoids in the = palliative=20 treatment of metastatic brain tumors. Cancer 18:298=96306, = 1965
Saadoun S, Papadopoulos MC, Davies DC, Krishna S, Bell = BA:=20 Aquaporin-4 expression is increased in oedematous human = brain=20 tumors. J Neurol Neurosurg Psychiatry 72:262=96265, 2002
Samoto K, Ikezaki K, Ono M, Shono T, Kohno K, Kuwano = M, et al:=20 Expression of vascular endothelial growth factor and its = possible=20 relation with neovascularization in human brain tumors. = Cancer Res=20 55:1189=961193, 1995
Sarntinoranont M, Chen X, Zhao J, Mareci TH: = Computational=20 model of interstitial transport in the spinal cord using = diffusion=20 tensor imaging. Ann Biomed Eng 34:1304=961321, 2006
Schuuring J, Bussink J, Bernsen HJ, Peeters W, van Der = Kogel=20 AJ: Irradiation combined with SU5416: microvascular = changes and=20 growth delay in a human xenograft glioblastoma tumor line. = Int J=20 Radiat Oncol Biol Phys 61:529=96534, 2005
Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey = VS, Dvorak=20 HF: Tumor cells secrete a vascular permeability factor = that=20 promotes accumulation of ascites fluid. Science 219: = 983=96985, 1983
Shapiro WR, Hiesiger EM, Cooney GA, Basler GA, = Lipschutz LE,=20 Posner JB: Temporal effects of dexamethasone on = blood-to-brain and=20 blood-to-tumor transport of 14C-alpha-aminoiso butyric = acid in rat=20 C6 glioma. J Neurooncol 8:197=96204, 1990
Shibata S: Ultrastructure of capillary walls in human = brain=20 tumors. Acta Neuropathol 78:561=96571, 1989
Shinonaga M, Chang CC, Suzuki N, Sato M, Kuwabara T:=20 Immunohistological evaluation of macrophage infiltrates in = brain=20 tumors. Correlation with peritumoral edema. J Neuro surg=20 68:259=96265, 1988
Shweiki D, Itin A, Soffer D, Keshet E: Vascular = endothelial=20 growth factor induced by hypoxia may mediate = hypoxia-initiated=20 angiogenesis. Nature 359:843=96845, 1992
Speirs V, Boyle-Walsh E, Fraser WD: Constitutive = co-expression=20 of estrogen and progesterone receptor mRNA in human = meningiomas by=20 RT-PCR and response of in vitro cell cultures to steroid = hormones.=20 Int J Cancer 72:714=96719, 1997
Stadlbauer A, Ganslandt O, Buslei R, Hammen T, Gruber = S, Moser=20 E, et al: Gliomas: histopathologic evaluation of changes = in=20 directionality and magnitude of water diffusion at=20 diffusion-tensor MR imaging. Radiology 240:803=96810, 2006
Stewart PA, Hayakawa K, Farrell CL, Del Maestro RF: = Quan=20 titative study of microvessel ultrastructure in human = peritumoral=20 brain tissue. Evidence for a blood-brain barrier defect. J = Neurosurg 67:697=96705, 1987
Stewart PA, Hayakawa K, Hayakawa E, Farrell CL, Del = Maestro=20 RF: A quantitative study of blood-brain barrier = permeability=20 ultrastructure in a new rat glioma model. Acta Neuropathol = 67:96=96102, 1985
Streffer JR, Bitzer M, Schabet M, Dichgans J, Weller = M: Re=20 sponse of radiochemotherapy-associated cerebral edema to a = phytotherapeutic agent, H15. Neurology 56:1219=961221, = 2001
Strugar J, Rothbart D, Harrington W, Criscuolo GR: = Vascular=20 permeability factor in brain metastases: correlation with=20 vasogenic brain edema and tumor angiogenesis. J Neurosurg = 81:=20 560=96566, 1994
Takamoto T, Sasaki M, Kuno T, Tamaki N: Flk-1 specific = kin ase=20 inhibitor (SU5416) inhibited the growth of GS-9L glioma in = rat=20 brain and prolonged the survival. Kobe J Med Sci 47: = 181=96191, 2001
Tjuvajev J, Kolesnikov Y, Joshi R, Sherinski J, = Koutcher L,=20 Zhou Y, et al: Anti-neoplastic properties of human cortico = tropin=20 releasing factor: involvement of the nitric oxide pathway. = In Vivo=20 12:1=9610, 1998
Uematsu H, Maeda M, Itoh H: Peritumoral brain edema in = intracranial meningiomas evaluated by dynamic = perfusion-weighted=20 MR imaging: a preliminary study. Eur Radiol 13: 758=96762, = 2003
Villalona-Calero MA, Eckardt J, Burris H, Kraynak M,=20 Fields-Jones S, Bazan C, et al: A phase I trial of human=20 corticotropin-releasing factor (hCRF) in patients with = peritumoral=20 brain edema. Ann Oncol 9:71=9677, 1998
Watanabe M, Tanaka R, Takeda N: Magnetic resonance = imaging and=20 histopathology of cerebral gliomas. Neuroradiology = 34:463=96469,=20 1992
Wei D, Wang L, He Y, Xiong HQ, Abbruzzese JL, Xie K: = Celecoxib=20 inhibits vascular endothelial growth factor expression in = and=20 reduces angiogenesis and metastasis of human pancreatic = cancer via=20 suppression of Sp1 transcription factor activity. Cancer = Res=20 64:2030=962038, 2004
Wrba E, Nehring V, Chang RC, Baethmann A, Reulen HJ, = Uhl E:=20 Quantitative analysis of brain edema resolution into the = cerebral=20 ventricles and subarachnoid space. Acta Neurochir Suppl=20 70:288=96290, 1997
Zangari M, Anaissie E, Stopeck A, Morimoto A, Tan N, = Lancet J,=20 et al: Phase II study of SU5416, a small molecule vascular = endothelial growth factor tyrosine kinase receptor = inhibitor, in=20 patients with refractory multiple myeloma. Clin Cancer Res = 10:88=9695, 2004

Abbreviation = Notes

BBB =3D blood=96brain barrier; COX-2 =3D cyclooxygenase-2; CT = =3D computed=20 tomography; DT =3D diffusion tensor; GBM =3D glioblastoma = multiforme; MR =3D=20 magnetic resonance; NO =3D nitric oxide; NOS =3D NO synthase; VEGF = =3D vascular=20 endothelial growth factor; VEGFR =3D VEGF receptor; WHO =3D World = Health=20 Organization.

Reprint Address

Walter Stummer, M.D., Department of Neurosurgery, University of = DFCsseldorf, Moorenstrasse 5, 40225 DFCsseldorf, Germany. email: = stummer@uni-duesseldorf.de= .=20

Walter Stummer, M.D., Department of = Neurosurgery,=20 DFCsseldorf, Germany