Cancer Research Meeting Calendar Sign up for Cancer Research eTOC's
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pathak, A. P.
Right arrow Articles by Bhujwalla, Z. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pathak, A. P.
Right arrow Articles by Bhujwalla, Z. M.
[Cancer Research 65, 1425-1432, February 15, 2005]
© 2005 American Association for Cancer Research

Cell and Tumor Biology

Characterizing Extravascular Fluid Transport of Macromolecules in the Tumor Interstitium by Magnetic Resonance Imaging

Arvind P. Pathak1, Dmitri Artemov1, B. Douglas Ward2, David G. Jackson3, Michal Neeman4 and Zaver M. Bhujwalla1

1 Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, Maryland; 2 Department of Biophysics, Medical College of Wisconsin, Milwaukee, Wisconsin; 3 MRC Human Immunology Unit, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Oxford, United Kingdom; and 4 Department of Biological Regulation, Weizmann Institute of Science, Rehovot, Israel

Requests for reprints: Zaver M. Bhujwalla, Department of Radiology, Johns Hopkins University School of Medicine, 208 Traylor Building, 720 Rutland Avenue, Baltimore, MD 21205. Phone: 410-955-9698; Fax: 410-614-1948; E-mail: zaver{at}mri.jhu.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Noninvasive imaging techniques to image and characterize delivery and transport of macromolecules through the extracellular matrix (ECM) and supporting stroma of a tumor are necessary to develop treatments that alter the porosity and integrity of the ECM for improved delivery of therapeutic agents and to understand factors which influence and control delivery, movement, and clearance of macromolecules. In this study, a noninvasive imaging technique was developed to characterize the delivery as well as interstitial transport of a macromolecular agent, albumin-GdDTPA, in the MCF-7 human breast cancer model in vivo, using magnetic resonance imaging. The transport parameters derived included vascular volume, permeability surface area product, macromolecular fluid exudate volume, and drainage and pooling rates. Immunohistochemical staining for the lymphatic endothelial marker LYVE-1 was done to determine the contribution of lymphatics to the macromolecular drainage. Distinct pooling and draining regions were detected in the tumors using magnetic resonance imaging. A few lymphatic vessels positively stained for LYVE-1 were also detected although these were primarily collapsed and tenuous suggesting that lymphatic drainage played a minimal role, and that the bulk of drainage was due to convective transport through the ECM in this tumor model.

Key Words: Interstitial drainage • macromolecules • breast cancer • lymphatic vessels • MRI


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The interstitial or extracellular matrix of the tumor presents a barrier against the transport of macromolecules and against cancer cell dissemination. The ability to image and characterize the transport of macromolecular agents through the extracellular matrix or tumor interstitium can be employed to understand factors affecting this transport, their role in macromolecular drug delivery and in cancer cell dissemination. Because the ease of transport of macromolecules through the matrix can be used as an index of its integrity, in vivo imaging can be used to design strategies for altering the porosity and integrity of the interstitial matrix.

The past few decades have seen extensive development of noninvasive imaging modalities for characterizing the tumor vasculature in animal models (1, 2) and in patients (3). However, there have been relatively few noninvasive imaging methods for characterizing macromolecular movement through the extracellular matrix. In normal tissue, macromolecules and cell debris are cleared from the interstitium through lymphatic drainage. In tumors, lymphatics have either been shown to be poorly developed (4) or nonfunctional due to the existing mechanical stresses (5), with the result that the transport of macromolecules and cell debris through the extracellular matrix are thought to occur primarily via convective flow (6, 7). Several recent studies have transformed our understanding of the role of this interstitial-lymphatic continuum of tumors, from a passive to a more active one (8–10). These include the role of interstitial fluid flow in remodeling the extracellular matrix (11), as well the meticulous molecular and functional characterization of the interstitial-lymphatic continuum (9, 12, 13).

Prior approaches for probing transport within the extravascular space, consisting of the interstitium and lymphatics (Fig. 1), have included (i) the use of fluorescence redistribution after photobleaching in situ (14), (ii) microlymphangiography (i.e., the injection of labeled macromolecules into the tumor interstitium and detection of its eventual uptake into the lymphatics), by multiphoton laser-scanning microscopy (4), (iii) micro-MR lymphangiography using dendrimer-based magnetic resonance imaging (MRI) contrast agents (15), and (iv) MRI using biotinylated BSA-GdDTPA (16).



View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Schematic illustrating compartmentalization of the macromolecular contrast agent albumin-GdDTPA. A, during the early phase of the MR protocol, the MMCA is confined to the vascular space, immediately followed by (B) extravasation from the hyper-permeable tumor vessels into the extravascular space, from which (C) it is cleared by convection and/or tumoral lymphatics, if present.

 
In this study, we have shown, for the first time, the ability of MRI of the macromolecular contrast agent (MMCA) albumin–gadolinium diethyltriamine penta-acetic acid (GdDTPA) to quantify delivery, transport rates, and volumes of macromolecular fluid flow through the interstitial-lymphatic continuum in tumors. Immunofluorescent microscopy of a lymphatic endothelial cell–specific marker LYVE-1 (lymphatic vessel endothelial hyaluronan receptor; ref. 17) together with a vascular endothelial cell–specific marker GSL-1 was done to determine the presence and contribution of lymphatic vessels to macromolecular drainage within the tumor. The MRI technique described here bridges the gap between traditional intravital lymphangiograpic approaches and multiphoton laser-scanning microscopy, by providing relatively high spatial resolution (100-250µm) for visualization of interstitial transport and drainage in deep-seated, optically nontransparent tumor tissue in vivo.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Tumor Model and Inoculations. MCF-7 cells were inoculated in the upper left thoracic mammary fat pad of five female severe combined immune deficient mice. Tumor cells were inoculated in a volume of 0.05 mL Hanks balanced solution (Sigma, St. Louis, MO) at a concentration of 106 cells/0.05 mL. All experimental animal protocols were approved by the Institutional Animal Care and Use Committee.

General Magnetic Resonance Imaging Paradigm. Mice were imaged 4 to 5 weeks postinoculation with tumor volumes in the range of 151.5 ± 111.7 mm3. MRI was done on a Bruker Avance 4.7-T instrument equipped with shielded gradients, using a custom-built RF volume coil placed around the body of the animal. Before placing each animal in the magnet, its tail vein was catheterized for administration of the MMCA albumin-GdDTPA. Because mice were imaged up to 140 minutes inside the magnet, an additional catheter containing the anesthetic solution was inserted s.c. for injecting additional anesthetic as required. One of the animals was imaged using biotinylated albumin-GdDTPA synthesized at the Weizmann Institute.

Multislice relaxation rate (T1–1) maps of the tumor were obtained using a saturation recovery method combined with fast-T1 SNAPSHOT FLASH imaging (flip angle = 10 degrees and echo time = 2 ms), as previously described (18). Images of six to eight slices (slice thickness of 1 mm) of the entire mouse cross-section including the tumor were acquired (128 x 128 matrix, 32-mm field of view, number of averages = 8) for three relaxation delays (100, 500, and 1,000 ms) for each slice. These MRI acquisition variables translate into an in-plane spatial resolution of 250 x 250 µm2. A multislice map of the completely relaxed magnetization (M0 map) was also acquired once at the beginning of the MR experiment, using a recovery delay of seven seconds. Images were acquired in two "phases" corresponding to the biphasic kinetics of the MMCA (Fig. 2). The first or "early phase" comprised of images obtained before i.v. administration of 0.2 mL of 60 mg/mL albumin Gd-DTPA in saline (dose of 500 mg/kg) and repeated every 7 minutes, starting at 3 minutes postinjection, up to 31 minutes (Fig. 2A). Because drainage of macromolecules in and around tumors either by convection or by the lymphatics is a slow event (19–21) with rates of 2.3 to 3.7 µL/g minutes (20), the second block of MR data was acquired up to 140 minutes postcontrast. This second block of acquisitions was classified as the "late-phase" of the MMCA, as it consists of late drainage events within the extravascular space of the tumor. At the end of the study, the mice were sacrificed and blood T1 values determined from blood samples taken from the tail vein (Fig. 2B).



View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. A, assuming negligible reflux of the contrast agent and fast exchange for the duration of the MR experiment, MMCA uptake was modeled as a linear function of time for the "early phase" (i.e., the first 31 minutes of the MR acquisition). For this phase, the slope (slope1) of the concentration-time curve provides the permeability surface area product, PS (µL/g min), and the y-intercept the vascular volume, VV (µ/g). During the early phase, we assume transport of MMCA from the vascular to the perivascular space is unidirectional, and that there is negligible transport (if any) of the MMCA from the perivascular to the drainage compartment. B, ninety to ~140 minutes after the administration of the MMCA (i.e., for the "late phase" of the MR experiment), MMCA uptake was also modeled as a linear function of time. During this phase, it was assumed there was unidirectional transport of the MMCA from the perivascular to the slow drainage compartment with a flux rate, FR (µL/g min) given by the slope (slope2) of the late phase concentration-time curve. Schematic plot of the MMCA concentration-time curve over the duration of the entire MR protocol of (C) all the vascular variables computed from the early phase of the experiment and (D) all the extravascular variables computed from the late phase of the experiment, respectively. The discontinuity on the time axis is to emphasize the 60-minute interval during which there was no MR acquisition.

 
Parameters describing the vascular and extravascular transport of the MMCA in MCF-7 tumors were calculated from the MMCA tissue concentration-time curve [{Delta}R1Tissue(t)] normalized to the MMCA concentration in the blood [{Delta}R1Blood(t)], by assuming three distinct compartments within the tumor (Fig. 2). Analogous to a model described by Bjornaes et al. (22), these three compartments are (i) the intravascular space, (ii) the perivascular region of the tumor interstitium that the MMCA initially extravasates into, and (iii) a more distant compartment consisting of regions of the interstitium within which slow macromolecular transport events such as convective and lymphatic drain occur. For both, the "early" and "late phases" of the MR experiment, we modeled the MMCA uptake as a linear function of time (Fig. 2C and D). We then did a voxel-wise multiple linear regression analysis to determine the regression variables for each phase of the MRI protocol as described in the following sections.

Calculation of Vascular Variables (Analysis of Early Phase). Assuming negligible reflux of the contrast agent and fast exchange for the duration of the MR experiment, MMCA uptake during the "early phase" (i.e., the first 31 minutes of the MR experiment) is given by:

(A)

Equation A is derived as follows:

If fractional volumes of individual compartments are VV, Vperi, Vdrain, and Vintracellular and the corresponding MMCA concentrations Cv, Cperi, Cdrain, and Cintracellular, respectively, we may write the total concentration as:

(B)

Because we use an MMCA that is excluded from the intracellular space, Cintracellular = 0 and C = CvVV + CperiVperi + CdrainVdrain. During the early phase of the experiment, we assume unidirectional transport of MMCA from the intravascular to the perivascular space and negligible transport (if any) of the MMCA from the perivascular to the drainage compartment (Fig. 2A). For the permeability surface area limited case, the rate of contrast uptake is limited by the PS of the vessel wall and is given by (23):

(C)

Solving Eq. C, gives:

(D)
where {alpha} is the constant of integration. Applying the initial condition C(0) = CvVV gives {alpha} = CvVV, which when substituted into Eq. D gives C(t) = PSCvt + CvVV or equivalently:

(E)
This is the same as Eq. A:

where [{Delta}R1Tissue(t)/{Delta}R1Blood(t)] is the ratio of the change in the relaxation rate of the tissue to that in the blood measured (by MRI) at some time t.

As shown by several investigators (18, 24), Eq. A indicates that the bulk of the T1 relaxation effect is initially proportional to the volume of the intravascular space since the contrast agent is confined to this space (Fig. 1A). The ratio of the change in relaxation rate of the tissue to that in the blood extrapolated to time t = 0 (i.e., the y-intercept) provides fractional vascular volume VV (µL/g) in that voxel (Fig. 2C). As the contrast agent extravasates into the perivascular space with time (Fig. 1B), the rate of change of the relaxation rate is proportional to the permeability surface area product; PS (µL/g min) for that vessel (i.e., the slope of the concentration-time curve) is equivalent to the permeability surface area product, PS (Fig. 2C). In other words, Eq. A is the equation of a straight line with slope PS and y-intercept VV (Fig. 2C). Voxel-wise multiple linear regression analysis of this phase yields spatial maps of tumor VV and PS. Only those voxels for which the individual model variables VV and PS passed the t test for significance (P ≤ 0.05) were considered for subsequent analysis.

Calculation of Bulk Macromolecular Contrast Agent Transport Variables (Analysis of Late Phase). Ninety to ~140 minutes after the administration of the MMCA (i.e., for the "late phase" of the MR experiment), MMCA uptake is given by Eq. F:

(F)

Equation F is derived as follows:

Approximately 90 to 140 minutes after the administration of the MMCA (i.e., for the late phase of the MR experiment), it is assumed there was unidirectional transport of the MMCA from the perivascular space to the slower drainage space with a transport rate, K2. For this phase, the rate of contrast uptake comprises of the continuing leakage (limited by the PS of the vessel) from the early phase as well as unidirectional transport from the perivascular to the drainage space with a flux rate, K2 given by:

(G)
Solving Eq. G gives:

(H)
where ß is the constant of integration. Applying the continuity condition that at the point of intersection (t = t') of the early- and late-phase concentration-time curves, Eqs. A and H must be equal, leads to PSCvt' + CvVV = PSCvt' + K2Cperit' + ß (i.e., ß = CvVV – K2Cperit'), which when substituted into Eq. H gives:

(I)
or equivalently:

(J)
This can be written as:

(K)
If , Eq. K reduces to . This is the same as Eq. F.

Equation F is also the equation of a straight line with slope FR and y-intercept ß2 (Fig. 2D). Both these variables can be determined from the concentration-time data using multiple linear regression analysis. If the flux rate, FR < 0 we refer to FR as the efflux rate, because a negative slope is representative of drainage (i.e., elimination of the contrast agent from the voxel). If the flux rate, FR > 0, we refer to FR as the influx rate, because a positive slope is representative of pooling of the contrast agent within the voxel. Unlike transport from the vascular to perivascular space, which is an intravoxel event, the nonvascular transport from the perivascular to the drainage space occurs between voxels (i.e., it is an intervoxel phenomenon). It is worth mentioning that (i) FR includes the continued extravasation of MMCA from the early phase as can be seen from Eq. K; (ii) during the early phase Cper = 0, and Eq. K reduces to Eq. A; and (iii) the y-intercept (ß2) of the late-phase concentration-time curve (Fig. 2D) cannot be employed for determining the volume of the slow drainage compartment analogous to the determination of VV in Eq. A, as K2 is unknown, and FR has a PS component (as can be seen from Eq. K). However, to further characterize this macromolecular transport in the interstitial compartment, we computed the apparent exudate volume, EVa (µL/g), by multiplying the flux rate (FR) by the duration of the late phase.

Kinetics of Albumin-GdDTPA in Blood. Because characterization of extravascular transport required the mice to be imaged up to {approx}140 minutes after i.v. administration of albumin Gd-DTPA, blood T1 levels were determined over this period to ascertain that intravascular T1–1 changes were not a source of changes in the MR signal in the tumor. In a separate study, described by us in ref. (25), blood T1 values were quantified over the entire duration of this protocol using a customized small RF solenoid tail-coil system.

Detection of Extravascular Macromolecular Drainage and Pooling. Using the multiple regression approach described above, both early- and late-phase model parameters were simultaneously computed from the concentration-time data. Voxels within each slice were further classified as either being (i) "pooling voxels" (i.e., voxels in which the MMCA accumulated over time) or (ii) "draining voxels" (i.e., voxels from which the MMCA was eliminated over time). Based on our knowledge of MMCA drainage kinetics, if MMCA was being drained extravascularly from a voxel either by convection or via lymphatics, one should observe a decreasing or negative slope for the late phase concentration-time curve corresponding to elimination of MMCA (Fig. 1C). In terms of the four model variables (VV, PS, ß2ß2 and FR), this is equivalent to the following condition: slope2 < slope1 or (slope2 – slope1) < 0 [i.e., (FR – PS) < 0]. In contrast, if the MMCA was pooling in the voxel, the slope of the late phase of concentration-time curve would be greater than that of the early phase [i.e., (slope2 – slope1) > 0, (FR – PS) > 0]. We tested for these unique conditions by subjecting the model parameters to the appropriate linear constraints during the multiple linear regression analysis by employing them in a general linear test for statistical significance. Consequently, only voxels for which the two conditions were statistically significant (P ≤ 0.05) were classified as "draining" or "pooling" voxels. We then examined the values of the corresponding variables VV, PS, FR, and EVa for both these classes of voxels. The exudate flux rate FR, was referred to as the "influx" rate for pooling voxels and as the "efflux" rate for draining voxels, respectively.

Not only does the inspection of the spatial distribution of these two classes of voxels allow us to make inferences about the pooling/drainage kinetics of the MMCA but by inspecting VV, PS, FR, and EVa for the same, we can begin to parameterize the interstitial-lymphatic space. All MR images were analyzed using the Analysis of Functional Neuro-Images program AFNI (26). Histologic analyses of immunofluorescent antibody stained sections obtained from the imaged tumors were done to identify the presence of lymphatic vessels.

Immunostaining. Adjacent 5-µm-thick formalin-fixed, paraffin-embedded tumor sections obtained at 500-µm intervals were cut onto silanized glass slides, cleared of paraffin in Histoclear II (National Diagnostics, Atlanta, GA) and rehydrated by means of graded alcohol baths. After rinsing in double-distilled water, slides were transferred to a preheated (95-100°C) buffer of 0.1 mol/L Tris-HCl (pH 9.0), 2 mmol/L EDTA and heated in a microwave for 10 minutes at 95°C to 100°C, and then cooled for 30 to 45 minutes. Slides were then blocked in PBS 5% FCS for at least 15 minutes.

Double Staining for Lymphatic Vessels and Blood Vessels. Lymphatic vessels were detected using an antibody to the lymphatic endothelial hyaluronan receptor LYVE-1 (17). Briefly, slides were incubated overnight at 4°C with 4.7 µg/mL of rabbit anti-mouse LYVE-1 (provided by Dr. David G. Jackson) in saline, rinsed (x3) in PBS and incubated with goat anti-rabbit Alexa-Fluor 488 (Molecular Probes, Inc., Eugene, OR) secondary antibody for 90 minutes at room temperature. For detection of blood vessel endothelia, slides were incubated with the isolectin, BS-I-B4 from Bandeiraea simplicifolia conjugated to biotin (Sigma), for 60 minutes at room temperature after another PBS rinse. Slides were then washed again in PBS before being incubated with an avidin-TRITC (tetramethylrhodamin-isothiocyanate) conjugate (Sigma) for 60 minutes at room temperature. Slides were counterstained with Hematoxylin-2 (Richard Allan Scientific, Kalamazoo, MI), dehydrated in graded alcohol baths and mounted in an aqueous mounting medium, Faramount (Dako Co., Carpenteria, CA). The presence of green fluorescence in the tissue sections was used to identify lymphatic endothelial cells of lymphatic vessels, whereas red fluorescence was used to identify regions containing vascular endothelial cells and blood vessels. Finally, for animals that were imaged using biotinylated albumin-GdDTPA, following LYVE-1 staining slides were incubated with streptavidin Marina Blue (Molecular Probes) secondary antibody for 90 minutes at room temperature, followed by biotin blocking (blocking kit, Vector Laboratories, Inc., Burlingame, CA), followed by the previous protocol for staining blood vessel endothelia and counterstaining.

Immunofluorescence Microscopy. All slides were viewed (at 40x) using a Nikon ECLIPSE-TS100 microscope (Nikon Instruments Inc., Melville, NY) equipped with Plan-fluor lenses and filters for detecting GFP/FITC, RFP/TRITC and UV/4',6-diamidino-2-phenylindole fluorescence. Regions of interest were digitized using a Nikon Coolpix 5000 digital camera (Nikon Co., Tokyo, Japan).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Representative concentration-time curves analyzed using the multiple regression analyses are shown in Fig. 3. The average values of vascular volume and permeability surface area product of the MCF-7 tumors was 4.60 ± 2.1 µL/g and 0.43 ± 0.19 µL/g min, respectively. Figure 3A illustrates a voxel for which (slope2 – slope1) > 0 [i.e., slope2 was significantly (P ≤ 0.05) greater than slope1]. Because such a concentration-time profile is representative of a voxel within which the MMCA is pooling, it was classified as a "pooling" voxel. Conversely, Fig. 3B illustrates a voxel for which (slope2 – slope1) < 0 [i.e., slope2 was significantly (P ≤ 0.05) smaller than slope1]. Such a concentration-time profile is representative of a voxel from which MMCA is being eliminated during the late phase and was classified as a "draining" voxel. This analysis, carried out on a voxel-wise basis, enabled us to examine the three-dimensional spatial distribution of pooling and draining regions within the tumor as shown in Fig 3C. Overall, there was a greater percentage of pooling voxels (19.9 ± 16.8) identified compared with draining voxels (2.9 ± 1.5; Fig. 3D), and pooling voxels were often found in proximity to draining voxels (Fig. 3C). In addition, a paired t test indicated that the influx rate (1.31 ± 0.28 µL/g min) was significantly (P = 0.001) greater than the efflux rate (–0.28 ± 0.17 µL/g min) for all the imaged tumors (Fig. 4A). Although the PS of the draining regions (0.87 ±0.54 µL/g min) tended to be higher than that of the pooling regions (0.25 ± 0.16 µL/g min; Fig. 4B), this difference was not significant. In contrast, the VV of the pooling regions (7.68 ± 1.55 µL/g) was significantly (P = 0.001) higher than that of the draining regions (3.68 ± 1.88 µL/g; Fig. 4C). Finally, there was also a significant difference (P = 0.001) in the apparent exudate volume of the pooling voxels (36.63 ± 7.70 µL/g) compared with the draining voxels (–7.90 ± 4.72 µL/g; Fig. 4D). All values presented are the mean ± SD. Figure 5 illustrates an example of interstitial drain occurring at the tumor-host tissue interface. Blood and lymphatic vessels were immunofluorescently labeled in the same tumor sections using a biotinylated primary antibody against lectin stained with avidin-TRITC, and LYVE-1 primary antibody stained with Alexa-Fluor 488, respectively. Detection of green fluorescence in these sections indicated the presence of lymphatic endothelial cells and lymphatics, whereas red fluorescence was observed in regions containing vascular endothelial cells and blood vessels. Each tumor section was also counter-stained with hematoxylin. Results from the immunohistochemistry studies obtained from a tumor are shown in Fig. 6. Lymphatic vessels exhibiting green fluorescence were observed in the tumor margins (Fig. 6A) with virtually no overlap with TRITC-stained blood vessels. Almost no intratumoral lymphatics were detected in the central regions of tumors. At high magnification (100x), the sparsely distributed lymphatic vessels at the tumor periphery were often seen to be intussuscepted by tumor cells (Fig. 6B-E). Figure 7 illustrates the intratumoral distribution of the biotinylated albumin-GdDTPA. Several MMCA-containing blood vessels are apparent in Fig. 7A, including areas wherein MMCA has extravasated. In contrast, no MMCA-containing lymphatic vessels were detected in any of the tumor sections.



View larger version (31K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. A, MMCA concentration-time curve typical of a pooling voxel along with regression lines for the early and late phases of the experiment. The discontinuity on the time axis is to emphasize the 60-minute interval during which there was no MR acquisition. B, MMCA concentration-time curve typical of a draining voxel along with regression lines for the early and late phases of the experiment. C, three-dimensional cutaway of a postcontrast MR image of a MCF-7 tumor overlaid with a functional map showing the spatial distribution of the extravascular draining and pooling voxels. D, comparison of the % total number of tumor voxels identified as either being draining or pooling, for all the animals. Columns, mean; bars, ±SD.

 


View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Summary of the extravascular variables determined for all the animals in this study. A, comparison of efflux and influx rates for draining and pooling voxels, respectively. B, comparison of the mean permeability surface area product for draining and pooling voxels. C, comparison of the mean vascular volume for draining and pooling voxels. D, comparison of the bulk exudate volume for draining and pooling voxels, respectively. Columns, mean; bars, ±SD.

 


View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. A, montage of slices from a MCF-7 tumor-bearing animal. M0 maps for each slice overlaid with a functional image in which one can see the spatial distribution of both draining (red) and pooling voxels (yellow). B, three-dimensional reconstruction of the same montage using cubic interpolation, cutaway to illustrate drainage (arrows) at the tumor-host tissue interface.

 


View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Photomicrograph obtained at 40x from a 5-µm-thick MCF-7 tumor section. A, lymphatic vessels exhibiting green fluorescence (LYVE-1 + Alexa-Fluor 488 stain) are visible in the tumor margin. B, high magnification (100x) H&E stained MCF-7 tumor section and (C) the same section stained with LYVE-1. D, another high magnification (100x) H&E stained MCF-7 tumor section and (E) the same section stained with LYVE-1. Arrows in each indicate lymphatic vessel lumen intussuscepted by tumor cells.

 


View larger version (64K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7. Photomicrograph obtained at 40x from a 5-µm-thick MCF-7 tumor section showing the intratumoral distribution of the MMCA albumin-GdDTPA (Streptavidin Marina Blue conjugate, blue). A, MMCA-bearing blood vessels (red) and (B) H&E stained section. C, magnified (100x) image of the region in A (gray square). Arrows, blood vessel lumen.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study, we have described a novel MRI paradigm to image and characterize delivery of MMCA through the vasculature and its transport within the tumor interstitium. We modeled the tumor to consist of (i) the intravascular space, (ii) the perivascular region of the tumor interstitium that the MMCA initially extravasates into, and (iii) the interstitium within which slow macromolecular transport events, such as convective and lymphatic drain, occur (21, 27). Several contrast-enhanced MRI studies have focused on the transport between (i) and (ii). Here for the first time we have derived quantitative variables characterizing extravascular events occurring within the interstitium.

The MMCA transport between compartments was modeled both, as a linear function of time and as consisting of an early phase and late phase. Our biphasic, multiple regression analysis approach was based on known dynamics of lymphatic-convective clearance of macromolecules within the tumor interstitium (16). Multiple linear regression analysis of each voxel was done to determine intra- and extravascular variables from the early and late phase of the MRI protocol, respectively. The advantage of employing a multiple linear regression approach is that it not only allows simultaneous computation of vascular and extravascular variables, but also permits easy identification of voxels based on any a priori chosen transport criteria. This is achieved by transforming the required criteria into a constraint for a general linear test to be tested during the linear regression.

Values of VV and PS derived from the early phase of the experiment were consistent with previously published data for this tumor model (18). Data from the late phase of the experiment were used to identify draining and pooling voxels and to quantify bulk volumes and rates associated with these voxels. Draining voxels exhibited significantly decreased enhancement, whereas pooling voxels exhibited significantly elevated MR signal enhancement during the late phase of the protocol, compared with the corresponding early phase. Pooling and draining voxels were often contiguous, with predominant drain occurring at or near the tumor-host tissue interface. Pooling regions exhibited higher EVa (i.e., volume of exudate) than draining regions, as well as a higher (in)flux rates compared with the draining voxels. These observations are consistent with the presence of a significant fraction of hyperpermeable vessels. These results are also consistent with those obtained in a study by Dafni et al. (16) in which VEGF165 over expression in C6 glioma tumors induced blood vessel hyperpermeability with an increased outward flux of contrast agent towards the tumor periphery. The total MR detectable volume of interstitial fluid an hour after the onset of the late phase can be approximated by the sum of the influx and efflux rates multiplied by 60 minutes [i.e., (1.31 + 0.28) x 60 {approx} 0.1 mL/h], which is comparable to the elimination rate for tumor interstitial fluid of ~0.2 mL/h reported by other investigators (20). In addition, the smaller exudate volume together with a lower (ef)flux rate of MMCA from draining voxels shows that the rate and amount of macromolecular clearance from the tumor interstitium is low. Factors such as elevated interstitial fluid pressure, blood vessel permeability, hydraulic conductivity, blood pressure, and lymphatic functionality affect the transport of macromolecules within the tumor interstitium (7); both convective and diffusive mechanisms (6) eventually drive MMCA drain within the tumor. Fluid exudate moves through the interstitial matrix along pathways of least resistance (28); elastin fibers within the extracellular matrix are thought to act as prelymphatic channels through the connective tissue, directing macromolecules into the lymphatics (Fig. 1; ref. 29). Our results suggest that clearance of macromolecules through the interstitial matrix was not efficient in this tumor model, consistent with the few lymphatic vessels detected in these tumors.

As shown by us in an earlier study (25), the small decrease in R1 (or 1/T1) of the blood measured over the duration of these experiments, ruled out changes in blood R1 as a potential source of the changes in the MR signal measured during the late phase of the experiment. Thus, the identification of both pooling and draining regions using the approach described here was most likely tracing a nonvascular or extravascular event, such as the movement of macromolecules into a lymphatic, or the movement of macromolecules via convective transport toward the tumor periphery. Furthermore, the time course of these observations is also consistent with previous reports of macromolecular transport characteristics through the tumor interstitium (16).

Although, the immunofluorescence studies revealed the presence of a small number of lymphatic vessels mostly localized to the tumor periphery, consistent with the drainage regions of interest identified by the MRI analyses, there was no colocalization of biotin-labeled MMCA and the LYVE-1 stain. The involvement of these peripheral lymphatics in MMCA clearance for this tumor type was most likely minimal. In addition, the low fractional area of LYVE-1 stained lymphatics in comparison to that of the draining regions of interest identified using MRI, implies that the bulk of the interstitial MMCA drainage for MCF-7 tumors was by convection and not via lymphatics.

The dynamic, noninvasive, in vivo functional assay described here, provides a powerful tool for probing the interstitial-lymphatic continuum and can be used for applications in which alterations in interstitial transport may occur. Such applications can include probing the remodeling of the extravascular space in tumors that often accompanies therapeutic interventions, and/or improving the delivery and transport of anticancer drugs.


   Acknowledgments
 
Grant support: National Cancer Institute grant RO1 CA90471.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Noriko Mori for synthesis of albumin-GdDTPA and Gary Cromwell for maintaining cell lines and inoculating the animals.

Received 10/14/04. Accepted 11/30/04.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Neeman M, Dafni H. Structural, functional, and molecular MR imaging of the microvasculature. Annu Rev Biomed Eng 2003;5:29–56.[CrossRef][Medline]
  2. Brasch R, Turetschek K. MRI characterization of tumors and grading angiogenesis using macromolecular contrast media: status report. Eur J Radiol 2000;34:148–55.[CrossRef][Medline]
  3. Padhani AR, Husband JE. Dynamic contrast-enhanced MRI studies in oncology with an emphasis on quantification, validation and human studies. Clin Radiol 2001;56:607–20.[CrossRef][Medline]
  4. Padera TP, Kadambi A, di Tomaso E, et al. Lymphatic metastasis in the absence of functional intratumor lymphatics. Science 2002;296:1883–6.[Abstract/Free Full Text]
  5. Padera TP, Stoll BR, Tooredman JB, Capen D, di Tomaso E, Jain RK. Pathology: cancer cells compress intratumour vessels. Nature 2004;427:695.[CrossRef][Medline]
  6. Jain RK. Transport of molecules across tumor vasculature. Cancer Metastasis Rev 1987;6:559–93.[CrossRef][Medline]
  7. Baxter LT, Jain RK. Transport of fluid and macromolecules in tumors. II. Role of heterogeneous perfusion and lymphatics. Microvasc Res 1990;40:246–63.[CrossRef][Medline]
  8. Detmar M, Hirakawa S. The formation of lymphatic vessels and its importance in the setting of malignancy. J Exp Med 2002;196:713–8.[Free Full Text]
  9. Oliver G, Detmar M. The rediscovery of the lymphatic system: old and new insights into the development and biological function of the lymphatic vasculature. Genes Dev 2002;16:773–83.[Free Full Text]
  10. Wiig H, Rubin K, Reed RK. New and active role of the interstitium in control of interstitial fluid pressure: potential therapeutic consequences. Acta Anaesthesiol Scand 2003;47:111–21.[CrossRef][Medline]
  11. Boardman, K. C. and Swartz, M. A. Interstitial flow as a guide for lymphangiogenesis. Circ Res 2003;92:801–8.[Abstract/Free Full Text]
  12. Pepper MS. Lymphangiogenesis and tumor metastasis: myth or reality? Clin Cancer Res 2001;7:462–8.[Abstract/Free Full Text]
  13. Jackson DG. The lymphatics revisited: new perspectives from the hyaluronan receptor LYVE-1. Trends Cardiovasc Med 2003;13:1–7.[CrossRef][Medline]
  14. Netti PA, Berk DA, Swartz MA, Grodzinsky AJ, Jain RK. Role of extracellular matrix assembly in interstitial transport in solid tumors. Cancer Res 2000;60:2497–503.[Abstract/Free Full Text]
  15. Kobayashi H, Kawamoto S, Sakai Y, et al. Lymphatic drainage imaging of breast cancer in mice by micro-magnetic resonance lymphangiography using a nano-size paramagnetic contrast agent. J Natl Cancer Inst 2004;96:703–8.[Abstract/Free Full Text]
  16. Dafni H, Landsman L, Schechter B, Kohen F, Neeman M. MRI and fluorescence microscopy of the acute vascular response to VEGF165: vasodilation, hyper-permeability and lymphatic uptake, followed by rapid inactivation of the growth factor. NMR Biomed 2002;15:120–31.[CrossRef][Medline]
  17. Banerji S, Ni J, Wang SX, et al. Jackson DG. LYVE-1, a new homologue of the CD44 glycoprotein, is a lymph-specific receptor for hyaluronan. J Cell Biol 1999;144:789–801.[Abstract/Free Full Text]
  18. Bhujwalla ZM, Artemov D, Natarajan K, Ackerstaff E, Solaiyappan M. Vascular differences detected by MRI for metastatic versus nonmetastatic breast and prostate cancer xenografts. Neoplasia 2001;3:143–53.[CrossRef][Medline]
  19. Aukland K, Nicolaysen G. Interstitial fluid volume: local regulatory mechanisms. Physiol Rev 1981;61:556–643.[Free Full Text]
  20. Butler TP, Grantham FH, Gullino PM. Bulk transfer of fluid in the interstitial compartment of mammary tumors. Cancer Res 1975;35:3084–8.[Abstract/Free Full Text]
  21. Dafni H, Israely T, Bhujwalla ZM, Benjamin LE, Neeman M. Overexpression of vascular endothelial growth factor 165 drives peritumor interstitial convection and induces lymphatic drain: magnetic resonance imaging, confocal microscopy, and histological tracking of triple-labeled albumin. Cancer Res 2002;62:6731–9.[Abstract/Free Full Text]
  22. Bjornaes I, Rofstad EK. Transvascular and interstitial transport of a 19 kDa linear molecule in human melanoma xenografts measured by contrast-enhanced magnetic resonance imaging. J Magn Reson Imaging 2001;14:608–16.[Medline]
  23. Tofts PS, Brix G, Buckley DL, et al. Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–32.[CrossRef][Medline]
  24. Roberts HC, Roberts TP, Brasch RC, Dillon WP. Quantitative measurement of microvascular permeability in human brain tumors achieved using dynamic contrast-enhanced MR imaging: correlation with histologic grade. AJNR 2000;21:891–9.[Abstract/Free Full Text]
  25. Pathak AP, Artemov D, Bhujwalla ZM. Novel system for determining contrast agent concentration in mouse blood in vivo. Magn Reson Med 2004;51:612–5.[Medline]
  26. Cox RW. AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res 1996;29:162–73.[CrossRef][Medline]
  27. Chary SR, Jain RK. Direct measurement of interstitial convection and diffusion of albumin in normal and neoplastic tissues by fluorescence photobleaching. Proc Natl Acad Sci U S A 1989;86:5385–9.[Abstract/Free Full Text]
  28. Ryan TJ. Structure and function of lymphatics. J Invest Dermatol 1989;93:18–24S.[CrossRef][Medline]
  29. Swartz MA. The physiology of the lymphatic system. Adv Drug Deliv Rev 2001;50:3–20.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 JNMHome page
P. T. Winnard Jr., A. P. Pathak, S. Dhara, S. Y. Cho, V. Raman, and M. G. Pomper
Molecular Imaging of Metastatic Potential
J. Nucl. Med., June 1, 2008; 49(Suppl_2): 96S - 112S.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
A. P. Pathak, D. Artemov, M. Neeman, and Z. M. Bhujwalla
Lymph node metastasis in breast cancer xenografts is associated with increased regions of extravascular drain, lymphatic vessel area, and invasive phenotype.
Cancer Res., May 15, 2006; 66(10): 5151 - 5158.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
D. W. Bartlett and M. E. Davis
Insights into the kinetics of siRNA-mediated gene silencing from live-cell and live-animal bioluminescent imaging
Nucleic Acids Res., January 12, 2006; 34(1): 322 - 333.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
H. Wiig, C. C. Gyenge, and O. Tenstad
The interstitial distribution of macromolecules in rat tumours is influenced by the negatively charged matrix components
J. Physiol., September 1, 2005; 567(2): 557 - 567.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pathak, A. P.
Right arrow Articles by Bhujwalla, Z. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pathak, A. P.
Right arrow Articles by Bhujwalla, Z. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online