Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print September 19, 2002, 10.1148/radiol.2252011605
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2252011605v1
225/2/527    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Michel, S. C. A.
Right arrow Articles by Kubik-Huch, R. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Michel, S. C. A.
Right arrow Articles by Kubik-Huch, R. A.
(Radiology 2002;225:527-536.)
© RSNA, 2002

Breast Imaging

Preoperative Breast Cancer Staging: MR Imaging of the Axilla with Ultrasmall Superparamagnetic Iron Oxide Enhancement1

Sven C. A. Michel, Dr med, Thomas M. Keller, Dr med, Johannes M. Fröhlich, Dr sc nat, Daniel Fink, PD Dr med, Rosmarie Caduff, PD Dr med, Burkhardt Seifert, PD Dr rer nat, Borut Marincek, Prof Dr med and Rahel A. Kubik-Huch, PD Dr med

1 From the Institute of Diagnostic Radiology (S.C.A.M., T.M.K., B.M., R.A.K.H.), Department of Obstetrics and Gynecology (D.F.), and Department of Pathology (R.C.), University Hospital Zurich, Switzerland; Department of Biostatistics, University of Zurich, Switzerland (B.S.); and Guerbet, Zurich, Switzerland (J.M.F.). From the 2001 RSNA scientific assembly. Received September 28, 2001; revision requested December 10; revision received January 7, 2002; accepted February 26. Supported in part by a grant from the Legat Frau Henriette Rossiez-Treichler, Zurich, Switzerland. Address correspondence to R.A.K.H., Institute of Diagnostic Radiology, Cantonal Hospital, CH-5404 Baden, Switzerland (e-mail: rahel.kubik@ksb.ch).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To evaluate magnetic resonance (MR) imaging with ultrasmall superparamagnetic iron oxide (USPIO) enhancement for preoperative axillary lymph node staging in patients with breast cancer by using histopathologic findings as the standard of reference.

MATERIALS AND METHODS: MR imaging was performed with a 1.5-T system within 24–36 hours after the start of intravenous slow-drip infusion of USPIO in 20 patients with breast cancer who were scheduled for surgery, followed by gadolinium-enhanced MR imaging. Lymph nodes were staged prospectively by using newly established criteria, and results were correlated with histologic findings.

RESULTS: In two patients, preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given; both patients were excluded from statistical analysis. Results of axillary staging with USPIO-enhanced MR imaging were true-positive in nine, true-negative in seven, false-positive in zero, and false-negative in two of 18 patients (sensitivity, 82%; specificity, 100%; positive predictive value, 100%; second reader, {kappa} = 1.0). Four hundred five lymph nodes were detected with MR imaging. For first and second readers, respectively, lymph node–based sensitivity was 83% and 73% and specificity was 96% and 97% ({kappa} = 0.68). USPIO as the intravascular contrast agent could not replace gadolinium for assessment of the primary tumor; however, no clinically relevant interaction was seen. Thus, an integrated imaging approach was feasible in all patients.

CONCLUSION: USPIO-enhanced MR imaging has the potential to become an adjunct to conventional MR imaging of the breast for preoperative assessment of axillary lymph nodes in patients with breast cancer.

© RSNA, 2002

Index terms: Breast neoplasms, MR 00.121411, 00.121412, 07.121411, 07.121412 • Breast neoplasms, metastases, 07.33 • Contrast media, comparative studies, 00.12143 • Iron • Lymphatic system, MR, 07.12143 • Lymphatic system, neoplasms, 07.33 • Magnetic resonance (MR), contrast enhancement, 07.12143


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Breast cancer has become the second most common cause of cancer-related deaths in women. Mammography and ultrasonography (US) are the current imaging modalities of choice for the initial assessment of suspicious lesions in the breast. The role of gadolinium-enhanced magnetic resonance (MR) imaging for preoperative assessment of the primary tumor is still evolving (2). When breast-sparing surgery is planned, MR imaging is advocated to exclude multifocal or multicentric carcinoma (1–3).

Treatment strategy and outcome in patients with breast cancer is dependent not only on the extent of the primary tumor but also on axillary lymph node staging, which plays an important role in the outcome of these patients. Lymph node status is one of the most important prognostic factors in breast cancer and is of particular value in choosing adjuvant therapy (4).

To be of value for clinical decision making, the accuracy of any imaging modality used to assess axillary nodes must closely match that of histopathologic findings. With the imaging techniques available today, however, which rely mainly on the size and shape of a lesion to differentiate benign from malignant nodes, preoperative assessment of axillary lymph node involvement is limited (5,6). Surgical lymph node dissection is thus used for exploration of the axilla. Although surgical dissection is an efficient diagnostic tool, it is accompanied by complications, including seroma formation, numbness, limitation of shoulder movement, and lymphedema (5,7).

About 60% of patients with breast cancer—usually those with small primary tumors—have negative results for cancer in the lymph nodes (8). In this subpopulation, complete axillary dissection may be regarded as too radical a treatment. It has been replaced in many institutions by less invasive surgical (ie, sentinel node lymphadenectomy [9]) or endoscopic (10–12) techniques. Recent data, on the other hand, demonstrate that patients with nodal involvement might benefit from preoperative adjuvant chemotherapy (13). Thus, a noninvasive imaging technique that allows accurate preoperative assessment of the axillary nodes and thus better preoperative patient selection would be of importance.

Dextran-coated ultrasmall superparamagnetic iron oxide (USPIO) particles have been introduced as a contrast agent for intravenous MR lymphography (14–16). In humans, the contrast agent remains, with an elimination half-life of 36 hours, in the intravascular compartment and is eventually incorporated into the reticuloendothelial system, thus providing information on lymph node morphology. Since lymph node macrophages—in contrast to tumor cells—can incorporate iron oxide nanoparticles, normal lymph nodes can be distinguished from metastases in normal-sized nodes on the basis of MR signal intensity characteristics (17). Findings in studies of patients with head and neck, urologic, and pelvic cancers have confirmed the potential for improved detection of lymph node metastases by using USPIO-enhanced MR imaging when compared with nonenhanced or gadolinium-enhanced MR imaging (18–20).

The aim of this prospective study was to evaluate USPIO-enhanced MR imaging for preoperative axillary lymph node staging in patients with breast cancer by using histopathologic findings as the standard of reference.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Patients
From April 2000 to April 2001, 20 (including three postmenopausal patients) consecutive female patients (mean age, 53 years ± 15 [SD]; age range, 22–76 years; mean height, 162 cm ± 5; height range, 155–170 cm; mean weight, 67 kg ± 9; weight range, 54–85 kg) who met the inclusion criteria and consented to participate were included in this prospective phase IIIb off-license study. According to the inclusion criteria, all patients had cytologically (n = 19) or histologically (n = 1) confirmed primary breast carcinoma and were scheduled for breast surgery, including axillary dissection. Exclusion criteria were a generally accepted contraindication for MR imaging, as well as a strong allergic disposition.

The institutional review board of the University Hospital Zurich approved the study, and written informed consent was obtained from all patients.

Contrast Agent
USPIO (ferumoxtrane, Combidex, Advanced Magnetics, Boston, Mass; and Sinerem, Guerbet, Roissy, France) was administered intravenously 24–36 hours prior to MR imaging as a slow drip infusion (2.6 mg of iron per kilogram of body weight, diluted in 100 mL of 0.9% saline solution) by using a flow rate of 2 mL/min for the first 10 minutes. When the agent was well tolerated (ie, no side effects such as rash or thoracic or lumbar pain occurred), the flow rate was increased to 4 mL/min for the rest of the infusion. Total infusion time was approximately 30 minutes.

No serious adverse events occurred in 22 examinations (two patients underwent follow-up USPIO-enhanced MR imaging of the axilla, see below) in 20 patients. In 18 (82%) examinations, no subjective or objective side effects were reported at all. In the remaining four (18%) examinations, mild side effects occurred less than 15 minutes after the start of infusion and subsided shortly after the end of infusion in three patients. Side effects were classified as moderate in one patient in whom symptoms had to be treated with an antihistamine. Symptoms (more than one symptom in some patients) included rash (9%; n = 2), pruritus (9%; n = 2), abdominal and/or lumbar pain (5%; n = 1), chest pain (14%; n = 3), and an orthostatic reaction (5%; n = 1). Lumbar and thoracic pain occurred immediately after infusion was started and was dependent on the flow rate. It subsided as soon as the flow rate was reduced.

MR Imaging
Imaging was performed with a 1.5-T MR system (Signa CV/i or Horizon; GE Medical Systems, Milwaukee, Wis). Initially, imaging of the axillary region was performed with the patient in the supine position by using a cardiac surface coil. After a localizer image was obtained, the following sequences were performed: transverse and coronal T1-weighted spin-echo (SE) 450/8 (repetition time msec/echo time msec) (matrix, 256 x 256; number of excitations, one for transverse images and four for coronal images), coronal T2-weighted fast SE 3,500–4,500/36 (matrix, 512 x 512; number of excitations, four; echo train length, 12) with and without fat saturation, and coronal gradient-echo (GRE) 120/15 (matrix, 256 x 256; flip angle, 25°) and coronal fast spoiled GRE 80/8 (matrix, 256 x 256; flip angle, 40°; section thickness, 3.5 mm; gap, 0.5 mm) sequences.

GRE sequences were tailored to improve susceptibility and T2* effect of USPIO (R2 at 0.47 T, 80–85 x 104 mol-1 · L · sec-1) (Sinerem [investigators’ brochure]. Roissy, France: Guerbet; 2000). A field of view of 22 x 22 cm and a section thickness of 3.5 mm with a 0.5-mm intersection gap were used for all sequences. Additional three-dimensional (3D) GRE 120/15 (matrix, 256 x 256; flip angle, 25°; section thickness, 2 mm) sequences were performed at the end of the protocol only when the patient agreed to remain in the imager for the additional time needed (12 of 20 patients).

Subsequently, the patient was examined in the prone position by using a dedicated bilateral breast surface coil. To investigate the intravascular properties of USPIOs resulting from their considerable T1 shortening effect (R1 at 0.47 T, 2.3 x 104 mol-1 · L · sec-1) (16), MR angiography (3D fast spoiled GRE time-of-flight 7.7/1.8; matrix, 256 x 192; flip angle, 45°; number of excitations, 0.5; field of view, 30 x 30 cm; section thickness, 3 mm) of the primary tumor was performed, followed by gadolinium-enhanced MR imaging of the breast. An unenhanced image was obtained, followed by four dynamic image acquisitions (3D fast spoiled GRE time-of-flight 7.7/1.8; matrix, 256 x 192; flip angle, 30°; number of excitations, one; field of view, 30 x 30 cm; section thickness, 3 mm) at 0.5, 1.0, 3.0, and 8.0 minutes after intravenous administration of contrast media (0.1 mmol/kg of gadopentetate dimeglumine, Magnevist; Schering, Berlin, Germany). The injection was followed by a saline flush. For the first and last enhanced cycle, all enhanced images were processed by subtracting the unenhanced images.

In lesions with early contrast material enhancement, a semiquantitative analysis of the signal intensity–to-time relationship was performed with the region-of-interest technique, as described previously (3). Five patients had already been referred for gadolinium-enhanced MR imaging of the breast prior (range, 1–22 days) to examination of the axilla. Sixteen of 20 patients underwent USPIO-enhanced MR angiography of the primary breast tumor.

Total imaging time for both examinations was less than 70 minutes. MR imaging of the axilla was performed in all 20 patients. Because of claustrophobia or back pain as a result of holding the same position during the examination, however, three patients wished to terminate the examination before acquisition of the USPIO- and gadolinium-enhanced MR images of the breast. In one patient, the tumor had been extirpated by means of stereotactically guided excisional biopsy before the MR examination. Gadolinium-enhanced (but not USPIO-enhanced) MR imaging of the breast was performed in this patient to exclude any residual or multifocal tumor.

Diagnostic image quality was achieved with the MR images of the breast and axilla in all patients except one, who was undergoing preoperative chemotherapy (see below). The images obtained at follow-up examination in that patient were hampered by motion artifacts.

Image Analysis
All axillary lymph nodes detected with MR imaging were evaluated before surgery on a console (Advantage Workstation 3.1; GE Medical Systems, Milwaukee, Wis). Mapping of lymph nodes was performed by means of comparison of two different section planes, and each lymph node was identified and numbered on the MR images and printed on film. The diagnosis was established in consensus fashion by a board-certified radiologist (R.A.K.H) and a resident experienced in MR imaging (S.C.A.M.).

While lymph node detection was easiest on GRE images, lymph node size was measured on T1-weighted SE images to minimize interference with susceptibility effects and thus overestimation of size. Criteria to distinguish benign from malignant nodes included size, shape, and USPIO uptake. A short-axis diameter exceeding 1 cm was considered indicative of a metastatic node, as was a round (ratio of longest to shortest axes smaller than 1.5) rather than an oval shape (21–23). Homogenous USPIO uptake of a lymph node on GRE and fast spoiled GRE images was considered indicative of benign nodes, whereas heterogeneous USPIO uptake or lack of uptake was considered indicative of malignant nodes (14).

On the basis of our previous experience, as well as results published in the literature (23,24), an additional criterion of USPIO uptake was introduced, since rim enhancement can be seen not only in malignant nodes but also in benign nodes because of the normal fatty hilum. Rim enhancement in an oval node with a smaller size on fat-saturated T2-weighted fast SE images than on GRE images was considered a sign of a benign node; rim enhancement in a round node with the same or larger size on fat-saturated T2-weighted images than on GRE images was regarded a sign of malignancy. By using these criteria, lymph nodes were classified as either benign or malignant.

Lymph nodes were classified into levels according to anatomic location: Level I indicated lower axillary nodes, located in the outer area of the chest, lateral to the pectoralis minor muscle; level II indicated midaxillary nodes, located beneath the pectoralis minor muscle; and level III indicated apical axillary nodes.

At the same time, for assessment of the primary tumor, the USPIO- and gadolinium-enhanced MR images of the breast were interpreted with regard to the delineation of the tumor, as well as the presence or absence of multifocal lesions, by a board-certified radiologist experienced in breast imaging (R.A.K.H.). On gadolinium-enhanced images, early (ie, 30 seconds after the start of contrast administration) and strong (>50% relative signal intensity increase) contrast enhancement on the dynamic T1-weighted 3D GRE images indicated malignancy (2,3).

Results of MR staging of the primary tumor (size, location, and presence of multifocality or multicentricity) and the axillary lymph nodes (number, size, and anatomic level of malignant nodes) were reported to the referring physician prior to surgery, and an outline of the lymph nodes was drawn and made available to the surgeon in the operating room.

The lymph nodes were reevaluated in all cases by an independent second reader (T.M.K.), who was unaware of the initial report, clinical findings, and definite histologic diagnosis. The lymph nodes were identified by using the lymph node mapping technique performed on GRE images, as documented on the MR images and printed on film. Characterization of the selected node was performed by comparing the enhancement and size features on all images.

Surgery and Histopathologic Correlation
Surgery was generally performed within 2 days ± 2 (range, 1–6 days) after MR imaging, except in two patients initially scheduled for surgery in whom preoperative chemotherapy was administered as a result of the findings of preoperative staging. These two patients were excluded from our statistical analysis. Nevertheless, in both patients, follow-up USPIO-enhanced MR imaging of the axilla and gadolinium-enhanced MR imaging of the breast were performed following chemotherapy and prior to the final treatment, which consisted of mastectomy and axillary dissection.

Of the remaining 18 patients, 12 underwent breast-sparing surgery and six underwent mastectomy. The decision to perform mastectomy was based on patient preference in two patients, tumor size in two patients, and multicentricity detected with MR imaging in one patient. In the remaining patient, breast-sparing surgery was performed initially, followed by subsequent mastectomy because of extensive in situ carcinoma.

Axillary dissection was performed in all patients by using a block resection of levels I and II of the ipsilateral side. For the purpose of histopathologic correlation, a lymph node was regarded as positive for malignancy when tumor cells were present at light microscopy, independent of the results of immunohistochemical staining. A node-to-node correlation between MR imaging and histologic findings was attempted; however, correlation was difficult for small nodes.

Statistical Analysis
Sensitivity, specificity, accuracy, and positive and negative predictive values were calculated on a disease (n = 18, malignant and benign cases on USPIO-enhanced MR images vs histopathologic findings) and lymph node basis.

In cases of discrepant numbers of lymph nodes counted on MR images and during surgery, the smaller number of detected nodes was defined as the total number in the patient. In cases in which node-to-node correlation was not feasible, nodes diagnosed as malignant on MR images were characterized as false-positive when they exceeded the number of histologically confirmed metastatic nodes in a particular patient. If more malignant nodes were detected with use of histologic findings than with MR imaging, they were regarded as false-negative.

{kappa} statistics were used to measure interobserver agreement. Strength of agreement was classified as slight ({kappa} = 0.20–0.40), moderate ({kappa} = 0.41–0.60), substantial ({kappa} = 0.61–0.80), or excellent ({kappa} = 0.81–1.0) (25,26). Interobserver {kappa} values for MR findings were calculated by using StatView 5.0 (SAS Institute, Cary, NC) and SPSS 9.0 (SPSS, Chicago, Ill) software on the basis of the detection of disease and the number of evaluated lymph nodes.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Preoperative Chemotherapy
In two patients, preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given before surgery. Both patients were thus excluded from our statistical analysis: A 37-year-old patient had been referred by her general practitioner to the department of gynecology because she had a palpable lump in her left breast and an invasive carcinoma had been cytologically confirmed. Since the breast parenchyma was dense, the tumor could not be delineated at mammography. MR images of the breast showed extensive tumor growth involving almost the entire left breast. USPIO-enhanced MR images demonstrated multiple metastatic axillary nodes that were subsequently confirmed by means of US-guided fine-needle aspiration biopsy. Follow-up MR images obtained after neoadjuvant chemotherapy showed no pathologic lesion in the breast and a regression of axillary nodes (Fig 1).



View larger version (101K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a. Cytologically confirmed invasive carcinoma in a 37-year-old patient with a palpable lump in the left breast. (a-c) Preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given before surgery. (a) Mediolateral mammographic images of both breasts. Since the breast parenchyma was dense, the tumor could not be delineated. R = right breast, L = left breast. (b) Axial gadolinium-enhanced T1-weighted MR image of the breast (early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at initial presentation showed extensive tumor growth (arrow). (c) Coronal USPIO-enhanced MR images obtained at initial presentation (left: T1-weighted SE 450/8; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) demonstrated multiple axillary nodes (arrows) mostly lacking USPIO enhancement that were subsequently confirmed to be metastatic by means of US-guided fine-needle aspiration biopsy. (d) Axial follow-up MR image obtained after neoadjuvant chemotherapy showed no contrast-enhanced lesion in the breast. (e) Coronal USPIO-enhanced MR images obtained after chemotherapy (left: T1-weighted SE; right: T1-weighted GRE) showed regression of axillary nodes; peripheral rim enhancement (arrows) surrounding the fatty hilum of the node is seen.

 


View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b. Cytologically confirmed invasive carcinoma in a 37-year-old patient with a palpable lump in the left breast. (a-c) Preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given before surgery. (a) Mediolateral mammographic images of both breasts. Since the breast parenchyma was dense, the tumor could not be delineated. R = right breast, L = left breast. (b) Axial gadolinium-enhanced T1-weighted MR image of the breast (early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at initial presentation showed extensive tumor growth (arrow). (c) Coronal USPIO-enhanced MR images obtained at initial presentation (left: T1-weighted SE 450/8; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) demonstrated multiple axillary nodes (arrows) mostly lacking USPIO enhancement that were subsequently confirmed to be metastatic by means of US-guided fine-needle aspiration biopsy. (d) Axial follow-up MR image obtained after neoadjuvant chemotherapy showed no contrast-enhanced lesion in the breast. (e) Coronal USPIO-enhanced MR images obtained after chemotherapy (left: T1-weighted SE; right: T1-weighted GRE) showed regression of axillary nodes; peripheral rim enhancement (arrows) surrounding the fatty hilum of the node is seen.

 


View larger version (100K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c. Cytologically confirmed invasive carcinoma in a 37-year-old patient with a palpable lump in the left breast. (a-c) Preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given before surgery. (a) Mediolateral mammographic images of both breasts. Since the breast parenchyma was dense, the tumor could not be delineated. R = right breast, L = left breast. (b) Axial gadolinium-enhanced T1-weighted MR image of the breast (early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at initial presentation showed extensive tumor growth (arrow). (c) Coronal USPIO-enhanced MR images obtained at initial presentation (left: T1-weighted SE 450/8; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) demonstrated multiple axillary nodes (arrows) mostly lacking USPIO enhancement that were subsequently confirmed to be metastatic by means of US-guided fine-needle aspiration biopsy. (d) Axial follow-up MR image obtained after neoadjuvant chemotherapy showed no contrast-enhanced lesion in the breast. (e) Coronal USPIO-enhanced MR images obtained after chemotherapy (left: T1-weighted SE; right: T1-weighted GRE) showed regression of axillary nodes; peripheral rim enhancement (arrows) surrounding the fatty hilum of the node is seen.

 


View larger version (69K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1d. Cytologically confirmed invasive carcinoma in a 37-year-old patient with a palpable lump in the left breast. (a-c) Preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given before surgery. (a) Mediolateral mammographic images of both breasts. Since the breast parenchyma was dense, the tumor could not be delineated. R = right breast, L = left breast. (b) Axial gadolinium-enhanced T1-weighted MR image of the breast (early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at initial presentation showed extensive tumor growth (arrow). (c) Coronal USPIO-enhanced MR images obtained at initial presentation (left: T1-weighted SE 450/8; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) demonstrated multiple axillary nodes (arrows) mostly lacking USPIO enhancement that were subsequently confirmed to be metastatic by means of US-guided fine-needle aspiration biopsy. (d) Axial follow-up MR image obtained after neoadjuvant chemotherapy showed no contrast-enhanced lesion in the breast. (e) Coronal USPIO-enhanced MR images obtained after chemotherapy (left: T1-weighted SE; right: T1-weighted GRE) showed regression of axillary nodes; peripheral rim enhancement (arrows) surrounding the fatty hilum of the node is seen.

 


View larger version (102K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1e. Cytologically confirmed invasive carcinoma in a 37-year-old patient with a palpable lump in the left breast. (a-c) Preoperative findings led to a change in therapeutic approach, and neoadjuvant chemotherapy was given before surgery. (a) Mediolateral mammographic images of both breasts. Since the breast parenchyma was dense, the tumor could not be delineated. R = right breast, L = left breast. (b) Axial gadolinium-enhanced T1-weighted MR image of the breast (early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at initial presentation showed extensive tumor growth (arrow). (c) Coronal USPIO-enhanced MR images obtained at initial presentation (left: T1-weighted SE 450/8; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) demonstrated multiple axillary nodes (arrows) mostly lacking USPIO enhancement that were subsequently confirmed to be metastatic by means of US-guided fine-needle aspiration biopsy. (d) Axial follow-up MR image obtained after neoadjuvant chemotherapy showed no contrast-enhanced lesion in the breast. (e) Coronal USPIO-enhanced MR images obtained after chemotherapy (left: T1-weighted SE; right: T1-weighted GRE) showed regression of axillary nodes; peripheral rim enhancement (arrows) surrounding the fatty hilum of the node is seen.

 
The second patient, a 42-year-old woman, had an extensive palpable lesion in the left upper outermost quadrant of the breast that was diagnosed by means of fine-needle aspiration biopsy as an invasive ductal carcinoma. The primary tumor was not seen with mammography, US, or gadolinium-enhanced MR imaging as a result of the specific microscopic growth pattern of the well-differentiated invasive ductal carcinoma. On USPIO-enhanced MR images, six of 27 detected axillary nodes in this patient were considered malignant, and metastatic involvement of the axilla was subsequently confirmed by means of fine-needle aspiration biopsy. On follow-up MR images obtained after chemotherapy, only one of 14 detected nodes was considered malignant, whereas histologic findings showed 14 of 18 nodes to be metastatic. However, image analysis was hampered in this case because of poor image quality due to motion artifacts.

Primary Tumor
In the remaining 18 patients, for whom a direct histopathologic correlation was available, gadolinium-enhanced MR images of the breast were acquired in 15. In 10 patients, gadolinium-enhanced MR imaging was performed in the same session as MR imaging of the axilla. Because of the lengthy presence of USPIO in the intravascular compartment, vessels within the breast parenchyma could be well delineated on the T1-weighted USPIO-enhanced MR images, which were acquired before gadolinium administration. Breast lesions, however, did not show any visible specific USPIO enhancement to improve lesion delineation.

On the other hand, USPIO did not show any clinically relevant interaction with the gadolinium enhancement properties of the primary breast tumor lesions and thus did not hamper image interpretation (Fig 2). Tumors were located in the left breast in four patients and in the right breast in 14. A stage pT1 tumor was diagnosed in 10 patients, and a stage pT2 tumor was diagnosed in seven patients. One patient had a histologically confirmed stage pT4 tumor. In 15 of the 18 patients undergoing surgery in whom MR imaging had been performed, the primary tumor was well detected and correctly diagnosed as malignant in all patients.



View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Primary tumor stage pT1pN1G3 in a 37-year-old patient. (a) USPIO-enhanced MR angiographic image (3D fast spoiled GRE time-of-flight 7.7/1.8; flip angle, 45°; maximum intensity projection acquired from all axial planes). (b) Gadolinium-enhanced MR image of the breast (axial early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at the same time as was the image of the axilla. While the tumor in the left breast shows early and strong peripheral enhancement after gadolinium administration (arrow in b), it cannot be delineated on the USPIO-enhanced image. Because of the lengthy presence of the USPIO in the intravascular compartment, vessels within the breast parenchyma are well seen (arrow in a). USPIO administration does not hamper interpretation of gadolinium-enhanced MR images.

 


View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Primary tumor stage pT1pN1G3 in a 37-year-old patient. (a) USPIO-enhanced MR angiographic image (3D fast spoiled GRE time-of-flight 7.7/1.8; flip angle, 45°; maximum intensity projection acquired from all axial planes). (b) Gadolinium-enhanced MR image of the breast (axial early subtraction 3D fast spoiled GRE 7.7/1.8 with a 30° flip angle) obtained at the same time as was the image of the axilla. While the tumor in the left breast shows early and strong peripheral enhancement after gadolinium administration (arrow in b), it cannot be delineated on the USPIO-enhanced image. Because of the lengthy presence of the USPIO in the intravascular compartment, vessels within the breast parenchyma are well seen (arrow in a). USPIO administration does not hamper interpretation of gadolinium-enhanced MR images.

 
Histologically confirmed multifocal lesions were diagnosed in five (28%) of 18 patients. In one patient with a multifocal lesion, no MR imaging of the primary tumor had been performed, and in another patient, the lesions had been removed by means of excision biopsy prior to MR examination. In the remaining three patients, MR imaging correctly depicted multifocal lesions.

Axillary Lymph Nodes
The ipsilateral axilla was negative for nodal malignancy in seven patients and positive in the remaining 11 patients. On the basis of the presence of nodal involvement, a sensitivity of 82%, specificity of 100%, accuracy of 89%, positive predictive value of 100%, and negative predictive value of 78% were achieved for both readers ({kappa} = 1.0) (Table).


View this table:
[in this window]
[in a new window]

  		Results of USPIO-enhanced MR Imaging of the Axilla by Using Histopathologic Findings as the Standard of Reference

 
According to assessment levels I, II, and III, a total of 405 lymph nodes were detected with MR imaging in the 18 included patients. At surgery, 319 nodes were removed in patients with level I and II nodes. During histologic investigation, three additional soft-tissue metastases were found in one patient, and an acanthoma was found in another. All four lesions had been detected previously with MR imaging and were excluded from statistical analysis. USPIO-enhanced MR imaging allowed characterization of 375 nodes (second reader, 372 nodes) as benign and 30 nodes (second reader, 33 nodes) as malignant (Figs 3–6). Of the 30 nodes characterized as malignant, seven had a short-axis diameter of 1 cm or less; of the 375 nodes characterized as benign, 34 exceeded 1 cm.



View larger version (48K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. Primary tumor stage pT1pN1(7/13)G3 in the same 37-year-old patient as in Figure 2. (a) Coronal USPIO-enhanced MR images (left: T1-weighted SE 450/8; middle: fat-saturated T2-weighted fast SE 4,500/36; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) show two round nodes (arrows) with sizes of 2.1 x 1.6 cm (upper node) and 2.0 x 1.5 cm (lower node). The nodes are hypointense on T1-weighted SE images and hyperintense on T2-weighted fast SE images and show only thin peripheral USPIO enhancement. (b) Photomicrograph of a histologic sample of one lymph node metastasis. Arrowheads point to small rims of normal lymphatic tissue. The lymph node is otherwise replaced by metastatic tissue. (Hematoxylin-eosin stain; original magnification, x3.)

 


View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. Primary tumor stage pT1pN1(7/13)G3 in the same 37-year-old patient as in Figure 2. (a) Coronal USPIO-enhanced MR images (left: T1-weighted SE 450/8; middle: fat-saturated T2-weighted fast SE 4,500/36; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) show two round nodes (arrows) with sizes of 2.1 x 1.6 cm (upper node) and 2.0 x 1.5 cm (lower node). The nodes are hypointense on T1-weighted SE images and hyperintense on T2-weighted fast SE images and show only thin peripheral USPIO enhancement. (b) Photomicrograph of a histologic sample of one lymph node metastasis. Arrowheads point to small rims of normal lymphatic tissue. The lymph node is otherwise replaced by metastatic tissue. (Hematoxylin-eosin stain; original magnification, x3.)

 


View larger version (51K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Primary tumor stage pT2N0G2 in a 61-year-old patient. Coronal USPIO-enhanced MR images (left: T1-weighted SE 450/8; middle: fat-saturated T2-weighted fast SE 4,500/36; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) shows an oval node (arrows) with a size of 2.1 x 1.0 cm. The node has a fatty hilum and demonstrates peripheral USPIO uptake. Because of the susceptibility artifacts of the contrast agent, the node appears to be larger on the GRE image.

 


View larger version (59K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. Primary tumor stage pT2N1G3 in a 39-year-old patient. (a) Coronal USPIO-enhanced MR images (left: T1-weighted SE 450/8; middle: fat-saturated T2-weighted fast SE 4,500/36; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) show a round node (arrowheads) with a size of 1.4 x 0.9 cm. The node is hyperintense on the T2-weighted image and shows only a small peripheral rim of USPIO enhancement. It was thus classified as malignant. An adjacent node (arrows) showed heterogeneous but strong USPIO enhancement and was falsely diagnosed as benign. (b) Photomicrograph of a histologic specimen of one metastatic lymph node. Arrowheads indicate small foci of metastatic tissue. Inset shows metastatic focus. (Hematoxylin-eosin stain; original magnification, x12; inset original magnification, x100.)

 


View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. Primary tumor stage pT2N1G3 in a 39-year-old patient. (a) Coronal USPIO-enhanced MR images (left: T1-weighted SE 450/8; middle: fat-saturated T2-weighted fast SE 4,500/36; right: T1-weighted fast spoiled GRE 80/8 with a 40° flip angle) show a round node (arrowheads) with a size of 1.4 x 0.9 cm. The node is hyperintense on the T2-weighted image and shows only a small peripheral rim of USPIO enhancement. It was thus classified as malignant. An adjacent node (arrows) showed heterogeneous but strong USPIO enhancement and was falsely diagnosed as benign. (b) Photomicrograph of a histologic specimen of one metastatic lymph node. Arrowheads indicate small foci of metastatic tissue. Inset shows metastatic focus. (Hematoxylin-eosin stain; original magnification, x12; inset original magnification, x100.)

 


View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Primary tumor stage pT2N1(1/25)G2 in a 42-year-old patient. Coronal USPIO-enhanced MR images (left: T2-weighted fast SE 4,500/36; right: T1-weighted GRE 120/15 with a 25° flip angle) show an oval node (arrows) with a size of 3.4 x 1.6 cm and lack of USPIO enhancement. The node was correctly classified as malignant. An adjacent small benign node (arrowheads) with homogenous USPIO enhancement is seen.

 
A node-to-node-based comparison between USPIO-enhanced MR imaging and histopathologic findings was not possible for small nodes because of block resection at surgery and the presence of few reference structures in the axilla. Most of those nodes were smaller than 4 mm and thus did not allow preoperative localization. In six patients, histologic findings showed more lymph nodes than did USPIO-enhanced MR images; in 12 patients, histologic findings showed fewer nodes. In those six patients whose histologic findings showed more lymph nodes, many were smaller than 2 mm. In the statistical analysis, 291 lymph nodes were included for the first reader and 296 for the second reader. Results are summarized in the Table.

Interobserver Agreement
Excellent interobserver agreement ({kappa} = 1.0) was achieved on the basis of the presence of disease (ie, metastatic involvement of the axilla). On a lymph node basis, substantial agreement was seen when the two patients undergoing preoperative chemotherapy were also included ({kappa} = 0.73), as well as when only those patients undergoing surgery without preoperative chemotherapy were included ({kappa} = 0.68; 405 lymph nodes in 18 patients).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
The integrated imaging strategy used in this study for the preoperative staging of breast cancer with USPIO-enhanced MR imaging of the axilla and dynamic gadolinium-enhanced MR imaging of the breast proved feasible and was well tolerated by all patients.

As was shown in this study, USPIO-enhanced MR imaging cannot replace dynamic gadolinium-enhanced MR imaging for the assessment of primary tumors in the breast. The small size and the specific coating of the USPIO nanoparticles lead to a prolonged blood half-life when compared with that of conventional iron oxide pharmaceuticals, since the particles are phagocytized less quickly by the reticuloendothelial system.

On the other hand, because of the size of the USPIO particles, they initially remain almost completely inside the vascular compartment. This makes the use of USPIOs favorable for MR angiography (17,27,28) and perfusion and permeability studies. This is corroborated by the findings in our study. Vessels in the breast parenchyma could be delineated clearly 24–36 hours after the start of contrast material injection. In contrast to gadolinium-enhanced imaging, however, no marked enhancement of the malignant lesion was seen, and thus, the lesion could not be better detected on USPIO-enhanced images. This finding was independent of both the time lap between the start of injection and imaging and the dose, as demonstrated in two patients (data not shown) in whom additional MR imaging of the primary tumor was performed immediately after USPIO infusion, with the same results. Since we did not perform imaging before USPIO administration, we were unable to investigate if there was any enhancement of the tumors, which seems likely.

Nevertheless, USPIO administration shows no clinically relevant interference with dynamic gadolinium-enhanced MR imaging of the breast. With one exception, all tumors in the present study were classified correctly as malignant on gadolinium-enhanced images by using the criteria described in the literature (3). In one patient with an extensive primary tumor, gadolinium-enhanced MR imaging, as well as mammography and US, failed to demonstrate the lesion. However, that finding can most likely be explained by the dissolute growth pattern of the tumor. Thus, both USPIO-enhanced MR imaging of the axilla and gadolinium-enhanced MR imaging of the primary tumor can be performed within the same imaging session, which is more convenient for the patient.

The finding that breast cancer shows less USPIO uptake than gadolinium uptake is not only of practical clinical relevance but also of theoretical interest. There are different hypotheses as to why breast tumors enhance on gadolinium-enhanced MR images: One theory is increased tumor microvasculature; another is increased tumor capillary permeability, and thus a higher diffusion and leakage of the contrast agent into the interstitium (29,30). Since, in the present study, the same tumors showed considerable enhancement after injection of gadolinium but to a much lesser extent after administration of the intravascularly distributed less permeable USPIO, our results favor the latter explanation. According to the findings in our study, it thus seems unlikely that—at least in patients with breast cancer—intravascular contrast agents will in general replace the widely used unspecific gadolinium compounds (31).

The results of our study, with a disease-based (ie, presence of nodal involvement) sensitivity of 82% and specificity of 100% for both readers and a lymph node–based sensitivity of 83% and specificity of 97% for the first reader, with substantial interobserver agreement, show the potential of USPIO-enhanced MR imaging as a diagnostic tool for preoperative nodal staging in patients with breast cancer. To our knowledge, this is the first study of USPIO-enhanced MR imaging for axillary nodal staging in patients with breast cancer. Yet, our results correlate well with those published in the literature on patients with head and neck, urologic, and pelvic cancers (18–20).

On the basis of the statistical exploitation of the phase II and phase III clinical results from Guerbet (B. Bonnemain, oral communication, August 1999), MR imaging was not performed before the administration of contrast material in our study. We were thus unable to assess signal intensity changes resulting from USPIO administration. According to previous results of the phase II and III studies, however, it seems unlikely that our results would have improved further if unenhanced images had been available.

While the sensitivity of USPIO-enhanced MR imaging of the axilla does not approach that of histopathologic findings, the integrated imaging strategy used in the present study can be of value for clinical decision making. As corroborated by the findings in our study, gadolinium-enhanced MR imaging of the breast may depict unsuspected multifocal, multicentric, or contralateral breast carcinoma and thus may lead to therapy changes (3).

With regard to nodal involvement, detection of additional nodes in a patient who already has enlarged nodes at the time of physical examination may not contribute much to patient care or surgical planning. However, when metastatic nodes are detected in a patient who should not have any malignant nodes according to current size criteria, therapy may indeed need to be changed. If USPIO-enhanced MR imaging depicts metastatic nodes in a patient scheduled for sentinel node lymphadenectomy, because of the high positive predictive value of the imaging test, it may be justified to replace this time-consuming surgical procedure with conventional surgical lymph node dissection—at least, if our preliminary results are confirmed in subsequent studies.

In patients with previously unexpected extensive lymph node involvement, this finding, along with the stage of the primary tumor, might result in a change of therapeutic approach, and neoadjuvant chemotherapy might be given (13,32). In our study, the therapy that was planned initially was changed because of preoperative findings in two of the 20 patients, and preoperative chemotherapy was applied. After neoadjuvant therapy, MR imaging was repeated, demonstrating the effect of therapy on tumor manifestations in the breast and axilla and thus serving as an "in vivo assay," allowing the therapeutic effect to be monitored. Further investigation is needed, however, to assess the effect of chemotherapy on MR findings.

In one of the two patients excluded from statistical analysis in our study, several small metastases were missed on the follow-up MR images obtained after chemotherapy. This might be expected, as difficulty in the interpretation of malignant nodes after chemotherapy is also a well-known problem with histologic findings (33–35).

Furthermore, whereas dynamic MR imaging of the breast was shown to be a valuable tool to determine tumor response in patients undergoing preoperative chemotherapy (13,32), results on gadolinium-enhanced MR images have to be interpreted with caution, since the change in gadolinium uptake behavior with chemotherapy can lead to an underestimation of tumor extent or even false-negative findings (34).

On the other hand, if dynamic gadolinium-enhanced MR imaging of the breast can help exclude multifocality and/or multicentricity in patients with dense breasts, which are difficult to interpret on mammograms, and if USPIO-enhanced MR images of the axilla show no malignant nodes, the surgeon may feel more confident about performing breast-sparing surgery, as well as sentinel lymph node dissection. The surgeon might also be assured that he or she will not miss lesions in the axillary nodes, especially since, even with an experienced surgeon, the sentinel node might be false-negative in about 3% of cases (36,37).

On the basis of our previous experience, as well as results in the literature (23,24), an additional criterion of USPIO uptake was introduced for lymph nodes with rim enhancement in addition to those criteria already published to refine the differential diagnosis of benign and malignant adenopathy (14). Peripheral enhancement can be seen not only in malignant nodes but also in benign normal-sized or enlarged reactive nodes, because the hilum of a lymph node consists of arteries, veins, lymphatic sinuses, and fatty tissue (23). In our study, rim enhancement in an oval node with a smaller size on fat-saturated T2-weighted images than on GRE images was considered a sign of a benign node, since the susceptibility effect of the iron oxide is more obvious on the latter image; conversely, rim enhancement in a round node with the same or larger size on fat-saturated T2-weighted images than on GRE images was regarded as a sign of malignancy.

Although we felt confident in assigning a correct diagnosis by using these criteria in the present study (in which we had substantial interobserver agreement), the interpretation of lymph nodes with rim or inhomogeneous enhancement remains challenging. Bellin et al (15) also found heterogeneity in the nodal architecture of some benign nodes in patients with pelvic and urologic tumors. This might be explained by the fact that USPIO particles accumulate in the medullary sinuses of the nodes, whereas the germinal centers and lymphoid follicles, which are largely devoid of macrophages, show no signal intensity change after USPIO injection, as shown in an animal study (38). The criteria used in our study thus need to be confirmed in a larger study population, as well as in studies of other anatomic regions.

There are some limitations to our study. We did not evaluate whether our results of axillary lymph node staging would have been different without the administration of USPIO. However, our results are superior to those of other studies in which lymph node staging was investigated with unenhanced or gadolinium-enhanced MR imaging (21,39–42). Furthermore, in our study, seven lymph nodes that would have been considered benign according to conventional criteria (ie, having a short-axis diameter <= 1 cm) were characterized as metastatic because of their USPIO uptake behavior. Thirty-four lymph nodes exceeding 1 cm were characterized as benign on the basis of USPIO enhancement.

As expected, because of the limited image resolution, small metastatic nodes (<2 mm) were missed with MR imaging. The clinical importance of the presence of a solitary micrometastasis for the prognosis of such a case is debatable (2,43–47). Furthermore, this problem might in part be overcome by future technical refinements and the introduction of new high-spatial-resolution sequences.

A node-to-node comparison between USPIO-enhanced MR imaging and histopathologic findings was not possible for small nodes because of the block resection and the few reference structures in the axilla. Most of the nodes were smaller than 4 mm and thus did not allow preoperative localization. In six patients, histologic findings showed more lymph nodes than did USPIO-enhanced MR imaging; in 12 patients, histologic findings showed fewer nodes. Pathologists usually find small nodes that are less than 2 mm in diameter and difficult to detect with any cross-sectional imaging modality because of partial volume effect. This could be an explanation for the discrepant results in the six patients in whom the number of nodes at pathologic investigation exceeded that at MR imaging (14).

On the other hand, in 12 patients, more lymph nodes were found at USPIO-enhanced MR imaging than during surgery. On the MR images, all detected lymph nodes were counted, whereas during surgery, only level I and II nodes were explored.

USPIO has to be administered within 24–36 hours prior to MR imaging, and thus, the patient needs two appointments in the radiology department; however, both can be scheduled on an outpatient basis. Although serious adverse events have been shown to be rare, and USPIO is generally well tolerated, side effects do occur.

In our study, adverse events occurring after USPIO administration were observed in 18% of examinations, which was within the expected range when compared with that of other prospective clinical studies (15,21). Symptoms of lumbar and thoracic pain depended on the flow rate, and they subsided immediately after the flow rate was reduced. This correlation has already been described for superparamagnetic iron oxides and has led to the recommendation of a biphasic administration scheme for such liver-specific agents (Endorem [leaflet]. Roissy, France: Guerbet; 2000) (Feridex [leaflet]. Wayne, NJ: Berlex; 2000).

In conclusion, USPIO-enhanced MR imaging has the potential to become an important adjunct to conventional MR imaging of the breast for preoperative assessment of axillary lymph nodes. USPIO as an intravascular contrast agent could not replace gadolinium for the assessment of the primary tumor. No clinically relevant enhancement interactions were seen, however, and thus, an integrated imaging approach that includes dynamic gadolinium-enhanced MR imaging of the breast was feasible in all patients within a reasonable amount of time.


   ACKNOWLEDGMENTS
 
We are indebted to Claudia Jenny for language editing.


   FOOTNOTES
 
Abbreviations: GRE = gradient echo, SE = spin echo, 3D = three dimensional, USPIO = ultrasmall superparamagnetic iron oxide

Author contributions: Guarantors of integrity of entire study, R.A.K.H., S.C.A.M., B.S.; study concepts and design, R.A.K.H., S.C.A.M.; literature research, S.C.A.M., R.A.K.H.; clinical studies, S.C.A.M., R.A.K.H., R.C.; data acquisition, T.M.K., S.C.A.M., R.A.K.H.; data analysis/interpretation, S.C.A.M., T.M.K., R.A.K.H., B.S.; statistical analysis, B.S., S.C.A.M., R.A.K.H.; manuscript preparation, definition of intellectual content, and editing, S.C.A.M., R.A.K.H.; manuscript revision/review and final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Gilbert FJ. New screening techniques for breast cancer (MRI). Dis Markers 1999; 15:115-116.[Medline]
  2. Friedrich M. MRI of the breast: state of the art. Eur Radiol 1998; 8:707-725.[CrossRef][Medline]
  3. Fischer U, Kopka L, Grabbe E. Breast carcinoma: effect of preoperative contrast-enhanced MR imaging on the therapeutic approach. Radiology 1999; 213:881-888.[Abstract/Free Full Text]
  4. Morrow M. Role of axillary dissection in breast cancer management. Ann Surg Oncol 1996; 3:233-234.[CrossRef][Medline]
  5. Pressman PI. Surgical treatment and lymphedema. Cancer 1998; 83(12 suppl American):2782-2787.[CrossRef][Medline]
  6. Noguchi M, Kurosumi M, Iwata H, et al. Clinical and pathologic factors predicting axillary lymph node involvement in breast cancer. Breast Cancer 2000; 7:114-123.[Medline]
  7. Yeoh EK, Denham JW, Davies SA, Spittle MF. Primary breast cancer: complications of axillary management. Acta Radiol Oncol 1986; 25:105-108.[Medline]
  8. Silverstein MJ, Skinner KA, Lomis TJ. Predicting axillary nodal positivity in 2282 patients with breast carcinoma. World J Surg 2001; 25:767-772.[CrossRef][Medline]
  9. Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391-398.[Medline]
  10. Kuhn T, Santjohanser C, Koretz K, Bohm W, Kreienberg R. Axilloscopy and endoscopic sentinel node detection in breast cancer patients. Surg Endosc 2000; 14:573-577.[CrossRef][Medline]
  11. Kuehn T, Santjohanser C, Grab D, Klauss W, Koretz K, Kreienberg R. Endoscopic axillary surgery in breast cancer. Br J Surg 2001; 88:698-703.[CrossRef][Medline]
  12. Malur S, Bechler J, Schneider A. Endoscopic axillary lymphadenectomy without prior liposuction in 100 patients with invasive breast cancer. Surg Laparosc Endosc Percutan Tech 2001; 11:38-41.[CrossRef][Medline]
  13. Aapro MS. Neoadjuvant therapy in breast cancer: can we define its role? Oncologist 2001; 6(suppl 3):36-39.[Abstract/Free Full Text]
  14. Anzai Y, Blackwell KE, Hirschowitz SL, et al. Initial clinical experience with dextran-coated superparamagnetic iron oxide for detection of lymph node metastases in patients with head and neck cancer. Radiology 1994; 192:709-715.[Abstract/Free Full Text]
  15. Bellin MF, Roy C, Kinkel K, et al. Lymph node metastases: safety and effectiveness of MR imaging with ultrasmall superparamagnetic iron oxide particles—initial clinical experience. Radiology 1998; 207:799-808.[Abstract/Free Full Text]
  16. Bellin MF, Beigelman C, Precetti-Morel S. Iron oxide-enhanced MR lymphography: initial experience. Eur J Radiol 2000; 34:257-264.[CrossRef][Medline]
  17. Weissleder R, Elizondo G, Wittenberg J, Rabito CA, Bengele HH, Josephson L. Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging. Radiology 1990; 175:489-493.[Abstract/Free Full Text]
  18. Anzai Y, McLachlan S, Morris M, Saxton R, Lufkin RB. Dextran-coated superparamagnetic iron oxide, an MR contrast agent for assessing lymph nodes in the head and neck. AJNR Am J Neuroradiol 1994; 15:87-94.[Abstract]
  19. Anzai Y, Prince MR. Iron oxide-enhanced MR lymphography: the evaluation of cervical lymph node metastases in head and neck cancer. J Magn Reson Imaging 1997; 7:75-81.[Medline]
  20. Harisinghani MG, Saini S, Slater GJ, Schnall MD, Rifkin MD. MR imaging of pelvic lymph nodes in primary pelvic carcinoma with ultrasmall superparamagnetic iron oxide (Combidex): preliminary observations. J Magn Reson Imaging 1997; 7:161-163.[Medline]
  21. Kvistad KA, Rydland J, Smethurst HB, Lundgren S, Fjosne HE, Haraldseth O. Axillary lymph node metastases in breast cancer: preoperative detection with dynamic contrast-enhanced MRI. Eur Radiol 2000; 10:1464-1471.[CrossRef][Medline]
  22. Yang WT, Lam WW, Yu MY, Cheung TH, Metreweli C. Comparison of dynamic helical CT and dynamic MR imaging in the evaluation of pelvic lymph nodes in cervical carcinoma. AJR Am J Roentgenol 2000; 175:759-766.[Abstract/Free Full Text]
  23. Uematsu T, Sano M, Homma K. In vitro high-resolution helical CT of small axillary lymph nodes in patients with breast cancer: correlation of CT and histology. AJR Am J Roentgenol 2001; 176:1069-1074.[Abstract/Free Full Text]
  24. Vasallo JL, Arreaza R, Munoz-Barragan L. Immunocytochemical study of vasopressin-like immunoreactive material in the gastrointestinal tract of the hedgehog Erinanceus europeus. Bol Asoc Med P R 1992; 84:67-69.[Medline]
  25. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977; 33:363-374.[CrossRef][Medline]
  26. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-174.[CrossRef][Medline]
  27. Loubeyre P, Zhao S, Canet E, Abidi H, Benderbous S, Revel D. Ultrasmall superparamagnetic iron oxide particles (AMI 227) as a blood pool contrast agent for MR angiography: experimental study in rabbits. J Magn Reson Imaging 1997; 7:958-962.[Medline]
  28. Engelbrecht MR, Saeed M, Wendland MF, Canet E, Oksendal AN, Higgins CB. Contrast-enhanced 3D-TOF MRA of peripheral vessels: intravascular versus extracellular MR contrast media. J Magn Reson Imaging 1998; 8:616-621.[Medline]
  29. Daldrup HE, Link TM, Blasius S, et al. Monitoring radiation-induced changes in bone marrow histopathology with ultra-small superparamagnetic iron oxide (USPIO)-enhanced MRI. J Magn Reson Imaging 1999; 9:643-652.[CrossRef][Medline]
  30. Schwickert HC, Stiskal M, Roberts TP, et al. Contrast-enhanced MR imaging assessment of tumor capillary permeability: effect of irradiation on delivery of chemotherapy. Radiology 1996; 198:893-898.[Abstract/Free Full Text]
  31. Turetschek K, Huber S, Floyd E, et al. MR imaging characterization of microvessels in experimental breast tumors by using a particulate contrast agent with histopathologic correlation. Radiology 2001; 218:562-569.[Abstract/Free Full Text]
  32. Tsuboi N, Ogawa Y, Inomata T, et al. Changes in the findings of dynamic MRI by preoperative CAF chemotherapy for patients with breast cancer of stage II and III: pathologic correlation. Oncol Rep 1999; 6:727-732.[Medline]
  33. Moll UM, Chumas J. Morphologic effects of neoadjuvant chemotherapy in locally advanced breast cancer. Pathol Res Pract 1997; 193:187-196.[Medline]
  34. Rieber A, Zeitler H, Rosenthal H, et al. MRI of breast cancer: influence of chemotherapy on sensitivity. Br J Radiol 1997; 70:452-458.[Abstract]
  35. Vinnicombe SJ, MacVicar AD, Guy RL, et al. Primary breast cancer: mammographic changes after neoadjuvant chemotherapy, with pathologic correlation. Radiology 1996; 198:333-340.[Abstract/Free Full Text]
  36. Veronesi U, Luini A, Galimberti V, Marchini S, Sacchini V, Rilke F. Extent of metastatic axillary involvement in 1446 cases of breast cancer. Eur J Surg Oncol 1990; 16:127-133.[Medline]
  37. Paganelli G. Sentinel node biopsy: role of nuclear medicine in conservative surgery of breast cancer. Eur J Nucl Med 1998; 25:99-100.[CrossRef][Medline]
  38. Lee AS, Weissleder R, Brady TJ, Wittenberg J. Lymph nodes: microstructural anatomy at MR imaging. Radiology 1991; 178:519-522.[Abstract/Free Full Text]
  39. Dooms GC, Hricak H, Moseley ME, Bottles K, Fisher M, Higgins CB. Characterization of lymphadenopathy by magnetic resonance relaxation times: preliminary results. Radiology 1985; 155:691-697.[Abstract/Free Full Text]
  40. Wiener JI, Chako AC, Merten CW, Gross S, Coffey EL, Stein HL. Breast and axillary tissue MR imaging: correlation of signal intensities and relaxation times with pathologic findings. Radiology 1986; 160:299-305.[Abstract/Free Full Text]
  41. Glazer GM, Orringer MB, Chenevert TL, et al. Mediastinal lymph nodes: relaxation time/pathologic correlation and implications in staging of lung cancer with MR imaging. Radiology 1988; 168:429-431.[Abstract/Free Full Text]
  42. Mumtaz H, Hall-Craggs MA, Davidson T, et al. Staging of symptomatic primary breast cancer with MR imaging. AJR Am J Roentgenol 1997; 169:417-424.[Abstract/Free Full Text]
  43. Attiyeh FF, Jensen M, Huvos AG, Fracchia A. Axillary micrometastasis and macrometastasis in carcinoma of the breast. Surg Gynecol Obstet 1977; 144:839-842.[Medline]
  44. Hartveit F, Lilleng PK. Breast cancer: two micrometastatic variants in the axilla that differ in prognosis. Histopathology 1996; 28:241-246.[CrossRef][Medline]
  45. Lilleng PK, Maehle BO, Hartveit F. The size of a micrometastasis in the axilla in breast cancer: a study of nodal tumor-load related to prognosis. Eur J Gynaecol Oncol 1998; 19:220-224.[Medline]
  46. Hoda SA, Chiu A, Prasad ML, Giri D, Hoda RS. Are microinvasion and micrometastasis in breast cancer mountains or molehills? Am J Surg 2000; 180:305-308.[CrossRef][Medline]
  47. Leong AS. The prognostic dilemma of nodal micrometastases in breast carcinoma. Gan To Kagaku Ryoho 2000; 27(suppl 2):315-320.



This article has been cited by other articles:


Home page
 RadiologyHome page
D. J. A. Margolis, J. M. Hoffman, R. J. Herfkens, R. B. Jeffrey, A. Quon, and S. S. Gambhir
Molecular Imaging Techniques in Body Imaging
Radiology, November 1, 2007; 245(2): 333 - 356.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
C. K. Kuhl
Current Status of Breast MR Imaging * Part 2. Clinical Applications
Radiology, September 1, 2007; 244(3): 672 - 691.
[Abstract] [Full Text]