| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic and Interventional Radiology OZ II, University
Hospital Essen, Hufelandstr. 55, 45122 Essen, Germany.
2 EPIX Medical, 71 Rogers St., Cambridge, MA 02142.
Received March 15, 2002; accepted after revision May 17, 2002.
Address correspondence to C. U. Herborn.
OBJECTIVE. The aim of our study was to evaluate the performance of a new blood-pool contrast agent, MS-325, in depicting regional lymph nodes when injected interstitially and in allowing the subsequent classification of the lymph nodes as normal or tumor-bearing (VX2 tumor).
MATERIALS AND METHODS. Six New Zealand white rabbits underwent adapted fast three-dimensional (3D) MR imaging before implantation of VX2 tumor cells in the flank and again 3 weeks after the implantation. For each imaging session, 0.5 mL of undiluted MS-325 was injected subcutaneously into both dorsal foot pads. For more than 120 min, the rabbits underwent repeated 3D MR imaging. The size of the individual lymph nodes and the amount of contrast agent uptake in the nodes were measured 5, 10, 15, 30, 60, and 120 min after the injection. After the rabbits had been sacrificed, their lymph nodes were removed and histopathologically analyzed.
RESULTS. In normal as well as tumor-bearing hindlegs, the subcutaneous administration of MS-325 resulted in rapid delineation of popliteal, inguinal, iliac, and paraaortal lymph nodes. Tumor invasion into lymph nodes presented as circumscribed signal voids in the areas infiltrated by tumor, whereas the surrounding residual lymphatic tissue showed enhancement identical to that of normal nodes.
CONCLUSION. In addition to providing a safe means of displaying the normal lymphatic system, MS-325—enhanced 3D MR lymphography depicts direct tumor invasion in lymph nodes.
The lymphatic system is of paramount importance in the dispersion of malignant tumors in that both continuous and metastatic invasion can occur along lymphatic vessels and nodes. Thus, prognosis and selection of the optimal therapy are heavily dependent on the determination of whether there is lymphatic system involvement. Current lymph-node staging strategies are based mostly on size criteria determined from cross-sectional images obtained by CT or MR imaging. The shortcomings of these criteria as evidenced by low sensitivity and specificity are well documented [1, 2]. Recently, new contrast agents based on monocristalline iron oxide nanoparticles as well as ultrasmall super-paramagnetic particles of iron oxide have been shown to accumulate in macrophages after IV administration. Accordingly, on T2-weighted MR images, these agents cause signal loss in normal lymph nodes [3, 4]. This selective enhancement allows some degree of differentiation between normal and tumor-bearing lymph nodes, especially in the pelvis. Problems associated with the visualization of negative enhancement—particularly in lymph nodes with partial tumor infiltration—appear to limit the clinical impact of this technique [5].
A more selective enhancement of the lymphatic system can be achieved with interstitial MR lymphography [6]. In one study, as many as five sequential lymph node groups were visualized in rabbits after subcutaneous injection of gadoterate meglumine, a conventional extracellular paramagnetic contrast agent [7]. Advantages of interstitial MR lymphography include vastly reduced doses of contrast agent, more selective accumulation in regional lymph nodes, and no harmful side effects. However, it still is unknown whether this approach allows differentiation between normal and tumor-bearing lymph nodes.
Beyond the commercially available extra-cellular paramagnetic agents, other gadolinium-based agents are now undergoing clinical testing. For example, MS-325 (EPIX Medical, Cambridge, MA), which incorporates novel chemical groups that permit highly effective reversible binding to albumin [8], is currently undergoing phase 3 clinical trials. A prolonged intravascular half-life induced by the binding to albumin renders MS-325 suitable as a blood-pool contrast agent [9, 10]. In addition, MS-325 was designed to be similar to approved paramagnetic agents in terms of stability, safety, and elimination profile. We hypothesized that the high content of albumin in lymphatic fluid would also make MS-325 useful for interstitial MR lymphography.
The purpose of our study was to evaluate the performance of MS-325 as an interstitial contrast agent for the depiction of regional lymph nodes and the accurate classification of the nodes as normal or tumor-invaded on the basis of three-dimensional (3D) data sets collected with an adapted 3D fat-saturated volumetric gradient-echo MR sequence in an animal model.
Materials and Methods
Contrast Agent
MS-325 is an extracellular, water-soluble, gadolinium-based MR contrast
agent. After IV injection, the agent strongly binds to albumin. This
reversible binding results in the retention of the agent in the blood pool.
This binding also enhances the magnetic properties of a single gadolinium ion
as much as 10-fold [11]. Thus,
even low doses of MS-325 yield vascular signal enhancement superior to that
provided by other contrast agents
[9,
10]. The pharmacokinetics and
elimination profile of MS-325, including vascular retention and renal
excretion of the material, render this compound suitable as a blood-pool
contrast agent. When extravasated into the interstitium, MS-325 causes no
inflammation [12], and this
quality coupled with its strong affinity for albumin make MS-325 attractive
for use in interstitial MR lymphography.
Animal Experiments
Experiments were conducted on six mature male New Zealand white rabbits
(weight range, 3.5-4.1 kg). The animal studies were approved by the
institutional review board for animal research. All experiments were performed
in accordance with state regulations governing the performance of animal
studies.
Each rabbit underwent interstitial MR lymphography both before and after implantation of a VX2 carcinoma in its left hindleg. The VX2 carcinoma (squamous cell carcinoma) model was chosen because it leads to invasive tumor growth with lymphatic metastasis [13, 14]. The highly concentrated VX2 tumor cell suspension (0.8-1.2 x 107 cells/mL) was injected 14 days after the initial examination. An 18-gauge needle was used to inject 1 mL of the cells in the subcutaneous tissues caudal to the popliteal lymph nodes of the left hindleg in all six rabbits. Before tumor cell implantation, we applied local anesthetic to the injection site.
Interstitial MR Lymphography
For interstitial MR lymphography, the rabbits were fully anesthetized with
a solution containing 10% ketamine hydrochloride (0.7 mL/kg Ketamin;
Sanofi-Ceva, Düsseldorf, Germany) and 2% xylazine hydrochloride (0.1
mL/kg Xylazin; Sanofi-Ceva). Interstitial MR lymphography of both hindlegs was
performed before tumor implantation and was repeated when the tumor had grown
to a palpable nodule with a diameter of approximately 3 cm. This growth was
reached between 21 and 28 days after the injection of the tumor cell
suspension.
The injection sites at both dorsal hind pads of each animal were carefully disinfected. Subsequently, 0.5 mL of undiluted MS-325 (0.25 mol/L) was administered subcutaneously into the inter-digital skin fold at the level of the metatarsal bones. The applied dose corresponded to approximately 0.03 mmol/kg of body weight. After the injection, the depot was gently massaged for approximately 30 sec to improve lymph drainage of the injection site [7]. Local tolerance at the injection site was evaluated 24 hr after the initial MR imaging examination.
All MR imaging was performed on a 1.5-T superconductive whole-body scanner (Magnetom Sonata; Siemens, Erlangen, Germany), using a dedicated transmit—receive head coil for optimized signal reception with the rabbits in the supine position. The conventional gradient system used had an amplitude of 40 mT/m and a slew rate of 200 mT/m per millisecond (rise time, 0.2 msec). To assess all phases of lymph node enhancement, we began image acquisition immediately after subcutaneous administration of the contrast material and repeated the imaging examination after 5, 10, 15, 30, 60, and 120 min.
A localizer sequence composed of three image stacks oriented in the coronal, sagittal, and axial planes was performed initially, and then a fast T1-weighted 3D gradient-echo sequence in the coronal plane (volumetric interpolated breath-hold examination [VIBE]) was used [15]. VIBE is performed with considerable fat suppression, which seems ideal for imaging lymph nodes because subtle lymphatic structures are often embedded in fat. On VIBE images, all tissues appear dark except those containing a substantial amount of T1-shortening contrast. Image parameters included a TR of 4.0 msec and an TE of 1.65 msec. The flip angle was 50°, with a field of view of 16-21 cm, an effective slice thickness of 0.64 mm, and a matrix size of 230 x 256. Postprocessing software and hardware (Virtuoso workstation; Siemens) provided interactive multiplanar viewing and 3D displays of the lymphatic structures when necessary. Maximum-intensity-projection images were postprocessed from the original data set.
Data Analysis
Three of the authors qualitatively assessed (via consensus) the enhancement
of the regional lymph nodes, lymphatic vessels, and distal parts of the
thoracic duct after the administration of MS-325. The size of each lymph node
was based on measurements obtained from the data set displaying the lymphatic
system to best advantage and was determined in three planes. Signal-to-noise
ratios (SNR) for each node were calculated by measuring signal intensities
(SI) in regions of interest encompassing the entire lymph node; the
calculation was performed using the usual equation: SNR = SIlymph
node/noise, in which noise was defined as the standard deviation from a
measurement of signal intensity within a circular 20-mm-diameter region of
interest outside the rabbit.
Regions of interest were placed by two of the authors, and measurements of signal intensity were based on source images. In regions with more than one lymph node (i.e., inguinal and iliac regions), SNR measurements were performed in the largest node as displayed on maximum-intensity-projection images. The size and position of the regions of interest remained unchanged for all image sets obtained during the same examination. The time-dependent enhancement profiles of the different lymph nodes were compared with regard to their anatomic location and their presence in a normal or tumor-bearing hindleg. Statistical analysis was based on a nonpaired Student's t test, using a p value at the 0.05 significance level.
Hypothesizing that MS-325-enhancement occurs only in normal lymphatic tissue, we characterized nodes in the tumor-bearing hindlegs displayed during the second MR lymphographic examination (performed 3 weeks after tumor induction) as either tumor-infiltrated or not. Furthermore, the degree of tumor infiltration was graded on the following scale (based on the amount of the lymph node volume involved): less than 25%, 25-75%, and more than 75%. The size of tumor-infiltrated lymph nodes was compared with the size of normal lymph nodes.
Histopathologic Analysis
After the second MR lymphographic examination, the tumor-bearing animals
were deeply anesthetized and then were sacrificed by exsanguination.
Popliteal, inguinal, and aortoiliac lymph nodes were exposed to determine
their positions and then were carefully removed. Lymph nodes proximal to the
paraaortic chain were not acquired because no correlation to MR images was
possible. The size of each lymph node was measured before the specimen was
fixed with a 10% formaldehyde solution and subsequent embedding in paraffin
for histopathologic processing. The histopathologic sections were obtained in
planes paralleling the MR imaging orientations. Serial sections of the entire
lymph node were obtained and subjected to H-and-E staining so that lymph node
morphology and tumor invasion could be assessed. Tumor infiltration was
quantified for each lymph node using the same scale as used in the analysis of
the MR images.
Results
MS-325—enhanced interstitial MR lymphography proved feasible in all the rabbits. In normal as well as tumor-bearing hindlegs, the subcutaneous administration of MS-325 into the dorsal foot pad resulted in rapid delineation of popliteal, inguinal, iliac, and paraaortal lymph nodes (Fig. 1A,1B). Within minutes after its interstitial injection, MS-325 drained into the lymphatic system. The popliteal lymph nodes became visible 5 min after the injection of the contrast agent, whereas the more proximal nodes showed contrast material uptake after 15 min. Efferent and afferent lymphatic vessels connecting lymph node groups were clearly visible.
|
|
Among the visualized lymphatic stations of the lower extremities, the popliteal lymph node group exhibited the highest signal intensity 5-15 min after subcutaneous MS-325 injection. The signal intensities in the inguinal lymph nodes group reached their signal intensity peak 15-30 min after the administration of MS-325. The iliac and paraaortal lymph nodes groups revealed the highest intensities approximately 30 min after contrast material administration (Fig. 2). Except for temporary swelling at the injection site, which lasted less than 24 hr, the local tolerance of MS-325 was excellent.
|
At the time of the second MR lymphographic examination, the subcutaneous VX2 tumors measured between 2.6 and 3.7 cm (mean, 3.3 cm) in diameter. Comparison of the speed of enhancement in tumor-bearing and normal hindlegs revealed no statistically significant difference (p > 0.05). Intraindividual comparison of lymph node contrast uptake revealed vast differences between the tumor-bearing and the normal hindlegs. Fifteen minutes after the injection, the mean SNR in the normal popliteal lymph nodes increased by a factor of 7.4, whereas SNR in invaded areas in the popliteal nodes of the tumor-bearing hindleg increased by a factor of only 2.3 (p < 0.05) (Fig. 3).
|
Tumor invasion presented as signal void in the areas with tumor infiltration; the surrounding residual lymphatic tissue revealed enhancement identical to that of normal lymph nodes (Fig. 4). Tumor infiltration of lymph nodes after the interstitial injection of MS-325 was evident in eight popliteal nodes in the tumor-bearing hindlegs of five rabbits. In three of these rabbits, four iliac nodes were also judged to be infiltrated. All other lymph node stations were considered tumor-free. Of the eight invaded popliteal nodes, five were considered to have tumor infiltration exceeding 75%, and three nodes were considered to have infiltration of less than 25%. Of the metastatic iliac nodes, two were considered to have 25-75% infiltration, and two nodes were deemed to contain less than 25% tumor cells.
|
At histopathology, nodal tumor infiltration was confirmed in five rabbits. In the remaining rabbit, no nodal tumor was revealed, although the VX2 tumor had a diameter of 2.9 cm. Histopathologic analysis revealed complete tumor invasion in two popliteal nodes and subtotal invasion (but > 75%) in three nodes. Metastases were smaller than 5 mm in the remaining three nodes, encompassing less than 25% of the nodal mass. Histopathology also confirmed tumor infiltration in four iliac lymph nodes. Two nodes had tumor infiltrates encompassing about half of their volume, whereas the remaining lesions, including the smallest metastasis measuring only 3 mm, occupied less than 25% of the nodal volume. Seventeen additional lymph nodes were confirmed as normal at histopathology, leading to an overall sensitivity and specificity for MR lymphography of 100% for both (Table 1). Size analysis of normal and metastatic nodes revealed a statistically significant difference in their axial diameters: the widths of the normal popliteal lymph nodes ranged from 0.6 to 0.9 cm (mean, 0.68 cm), whereas the axial diameters of metastatic lymph nodes ranged from 0.8 to 1.2 cm (mean, 0.92 cm) (p < 0.05) (Fig. 5A,5B). The differences in craniocaudal size did not prove statistically significant (mean, 1.1 cm vs 1.3 cm) (p > 0.05). Nodal size correlated well with postmortem measurements of tumor-bearing nodes in the rabbits (R = 0.87).
|
|
|
Discussion
MS-325—enhanced interstitial MR lymphography is feasible and permits differentiation of normal from metastatic lymph nodes in the animal model used in our study. Because of the affinity of MS-325 to albumin, this interstitially administered agent is rapidly taken up by the lymphatic system. Marked T1-shortening associated with the agent allows the entire lymphatic system to be displayed, with depiction of both lymph vessels and lymph nodes from the feet to the chest. Enhancement is limited to normal lymphatic tissues, so the technique described permits differentiation of normal and tumor-infiltrated lymph nodes. Further investigation is required to identify the optimal administered dose and to determine the efficacy of MS-325—enhanced interstitial MR lymphography in humans.
Diligent evaluation of the lymphatic system is pivotal to deciding the optimal therapeutic management for most forms of malignant disease. Although CT and sonography can aid in the assessment of lymph node size and location, these modalities often fail to provide data needed to evaluate the parenchymal structure of lymphatic tissue within nodes [16]. As with CT, MR imaging to date has been used mainly to localize and measure the size of lymph nodes [17]. Image acquisition in the coronal plane has proven useful in this context, particularly in the retroperitoneum [18]. However, inherent T1- and T2-contrast mechanisms have not been helpful for use in characterizing lymph nodes.
To characterize lymph nodes as normal or metastatic, researchers can exploit the differences in the tissue-specific contrast agent uptake. Encouraging results have been reported after IV administration of ultrasmall super-paramagnetic iron oxide nanoparticles. Within 12 hr after administration of the particles, phagocytosis by reticuloendothelial cells results in the accumulation of these particles in the medullary sinuses of lymph nodes [19]. The T2-shortening caused by the iron particles is limited to normal lymph nodes. Hence, normal nodes are displayed as areas of susceptibility-mediated decreased signal on T2-weighted MR images [5]. Metastatic lymph nodes, on the other hand, do not take up particles and thus maintain their inherent signal characteristics [20]. In spite of encouraging results in staging lymph nodes in the pelvis, clinical use of this technique is limited by the rather extended pharmacologic half-life of the iron oxide nanoparticles (which can be as long as 4 days) [21], the unpredictability of iron-induced susceptibility artifacts, and a notoriously heterogeneous enhancement profile in normal lymph nodes [22]. Several of these limitations might be overcome with recently developed T1-shortening contrast agents, including polymeric gadolinium compounds [23, 24] and liposome-encapsulated gadubutrol [25, 26]. At this time, however, the use of these contrast agents is still at the preclinical stage.
Compared with MR imaging using an IV contrast material, interstitial MR lymphography permits a more selective and more comprehensive display of the lymphatic system. Thus, a small dose of 0.03 mmol/kg of MS-325 (0.5 mL) injected subcutaneously into the dorsal foot pads of the rabbits was sufficient for striking contrast enhancement of the draining lymphatic vasculature and successive lymph node groups. Maximum-intensity-projection images provided a comprehensive display of the entire lymph system—both nodal structures as well as afferent and efferent lymphatic vessels in the hindlegs and the abdomen (Fig. 1A,1B). The applied dose resulted in homogeneous enhancement of normal lymph nodes. Although the signal was uniform in the early phase, nodal MS-325 uptake is likely to vary between paracortical and medullary regions. Tagged probes may elucidate such distribution patterns in future studies.
The feasibility of interstitial MR lymphography has been successfully shown with commercially available extracellular paramagnetic agents even in humans [27]. In one study, an interstitially administered extracellular contrast agent lacking the protein-binding affinity inherent to MS-325 was taken up in large part by the capillary system, reducing the ability to delineate the lymphatic system. Our data show MS-325 to be a far better contrast agent for interstitial MR lymphography. The agent has successfully passed phase 1 and phase 2 clinical evaluations as an IV blood-pool contrast agent [28]. The albumin-binding affinity that ensures the prolonged presence of the agent in the vascular system is also responsible for its suitability as an agent for interstitial MR lymphography.
The interstitial application of MS-325 seems safe. Relating to the inertness of this contrast agent, murine experiments have shown that extravasated MS-325 is not associated with inflammatory reactions in the subcutaneous tissues [12, 29]. Furthermore, the applied volume required is extremely low. Interpolated to human standard weights, the MS-325 volume required for interstitial MR lymphography would be 5-10 mL. The sevenfold SNR increase that we found on T1-weighted MR images suggests a potential for considerable dose reduction. Even these volumes should be well tolerated, however. When combined with a local anesthetic, the subcutaneous injection of 5 mL of gadoterate meglumine in the dorsum of the foot has been shown to cause little or no discomfort [27].
For display of the MS-325 uptake in the lymphatic system, VIBE was valuable because of its considerable fat suppression and sensitivity for T1-shortening compounds. After the interstitial administration of MS-325, the technique permitted accurate delineation of the lymphatic structures. The inherent 3D quality of VIBE aided in the morphologic analysis of individual lymph nodes.
The lymphatic vasculature together with popliteal and inguinal lymph nodes markedly enhanced within 15 min after the subcutaneous MS-325 administration. The aortoiliac lymph nodes and inferior parts of the thoracic duct were best visualized 30 min after injection of the compound. Washout of MS-325 occurred quickly and was comparable to the time required for washout of gadoterate meglumine [7], gadofluorine, and gadofluoramide [25], with signal intensities reaching almost baseline values within 2 hr after the injection.
The exact means of MS-325 uptake into the lymphatic system have not been determined. The process is likely driven by a combination of osmosis and pressure [30]. Lack of enhancement of lymphoid tissue in a node after the subcutaneous injection of MS-325 corresponded to the presence of tumor infiltration by VX2 carcinoma in 12 popliteal and iliac lymph nodes. At the same time, the enhancement of normal lymph nodes was remarkably homogeneous. With this technique, metastases as small as 3 mm were readily detected. Compared with normal nodes, tumor-invaded lymph nodes appeared a bit larger on axial MR images. Differences were statistically significant only in the axial plane, and considerable overlap was observed.
Virtually all lymph nodes exhibited some peripheral enhancement, which represented residual lymphoid tissue in the node. Our tumor model has been used in previous studies evaluating the lymphatic system with different contrast agents and has reliably metastasized to nearby lymph nodes. In our study, the determination of tumor invasion of lymph nodes was based on peak-enhancement MR images. In later phases, washout caused signal inhomogeneities, pointing to a potential limitation of this method. Physiologic washout in the late phase might result in the misinterpretation of normal lymph nodes as tumor-invaded.
Clearly, more studies of the lympholitic properties of MS-325 are warranted. At this time, the potential of interstitial MS-325—enhanced MR lymphography seems great because the technique promises not only to display the entire lymphatic system but also to permit the differentiation of normal lymph nodes from those infiltrated by tumor.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Lohrmann, E. Foeldi, O. Speck, and M. Langer High-resolution MR lymphangiography in patients with primary and secondary lymphedema. Am. J. Roentgenol., August 1, 2006; 187(2): 556 - 561. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
)
enhanced 5-15 min after injection, more proximal lymph node groups ([UNK] =
iguinal nodes; 
= iliac and aortal nodes) reached maximal signal
intensity 15-30 min after administration of MS-325. Horizontal lines above
bars represent standard error of mean.
