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


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 Weinstein, D.
Right arrow Articles by Degani, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Weinstein, D.
Right arrow Articles by Degani, H.
(Radiology. 1999;210:233-240.)
© RSNA, 1999

Technical Developments

Breast Fibroadenoma: Mapping of Pathophysiologic Features with Three-Time-Point, Contrast-enhanced MR Imaging—Pilot Study

Daphna Weinstein, MD2,1, Shalom Strano, MD3,2, Patricia Cohen, MD3, Scott Fields, MD4, J. Moshe Gomori, MD4 and Hadassa Degani, PhD2

1 Department of Biological Regulation, Weizmann Institute of Science, Rehovot, 76100, Israel (D.W., H.D.)
2 Department of Radiology, Kaplan Hospital, Rehovot, Israel (S.S.)
3 Departments of Pathology (P.C.)
4 Radiology (S.F., J.M.G.), Hadassah Medical Center, Jerusalem, Israel.


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Appendix
 References
 
The capability of three-time-point, gadolinium-enhanced magnetic resonance imaging to depict vascular permeability and extracellular volume fraction of breast fibroadenoma was evaluated with histopathologic correlation. This method demonstrated an even distribution of high extracellular volume fraction and low to moderate microvascular permeability in these common breast lesions, providing a nonsurgical means of improving the accuracy of diagnosis of fibroadenoma.

Index terms: Breast neoplasms, MR, 00.121412, 00.311 • Magnetic resonance (MR), tissue characterization, 00.121412, 00.311


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Appendix
 References
 
A major goal in breast diagnosis is to be able to differentiate between benign and malignant lesions in a noninvasive and reliable manner (1–4). Fibroadenoma, which consists of both fibrous and glandular tissue, is the most common benign tumor of the female breast. It occurs at any age in the reproductive period, but is more common before age 30 years. The lesion is usually solitary, but rarely may be multiple and bilateral. It usually appears as a discrete, rounded, lobulated, freely movable nodule in the breast, and varies in size from less than 1 cm to giant forms that are 10–15 cm in diameter. In macroscopic specimens, fibroadenomas are typically firm, rubbery, and grayish white. Histopathologically, they are seen as delicate, fibroblastic stroma enclosing glandular and cystic spaces lined with two layers of luminal epithelial cells and outer myoepithelial cells. Hormone-related changes can cause a slight increase in size during the late phase of the menstrual cycle and during pregnancy. Postmenopausal changes may result in regression, calcification, or both (5,6).

Traditionally, fibroadenomas have been divided into two types on the basis of the arrangement of stroma relative to the epithelial structures: (a) pericanalicular fibroadenoma, which consist of intact, round to oval gland spaces lined by single or multiple layers of cells, with concentric arrangement of the adjacent stroma; and (b) intracanalicular fibroadenoma, in which the stroma has undergone more active proliferation with compression of the gland spaces and, as a consequence, the glandular lumina are compressed into slitlike, irregular clefts, and the epithelial lining appears as narrow strands in the fibrous stroma. These two patterns can coexist in the same tumor. There may also be several forms of stromal changes: Mucinous or myxoid change is frequent, whereas marked hyalinization or calcification can be seen in older patients. Carcinoma, either invasive or noninvasive, is a rare occurrence in fibroadenoma. Traditionally, the risk for subsequent carcinoma in patients with fibroadenoma has not been considered to be higher than that in the general population (7).

Regarding the vascularity of fibroadenomas, it was found that fibroadenomas exhibited a lower vessel density than did recurrent invasive breast cancer lesions (8). Findings at color Doppler ultrasonography (US) showed that blood flow is lower in fibroadenomas than in breast cancer lesions (9,10).

Gadolinium-enhanced magnetic resonance (MR) imaging has been applied to improve breast cancer detection and diagnosis (11,12). Most invasive breast carcinomas enhance profoundly with gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany). However, the benign lesions, including fibroadenomas, enhance to various levels from minimal to intense (13–17). Several approaches were applied to overcome the problem of enhancement of benign and malignant lesions (14–17). In cases of fibroadenoma, architectural changes found on T2-weighted images and on contrast material–enhanced images were found to be useful in the demonstration of features confirmed histopathologically (17,18).

Most important, dynamic contrast-enhanced MR imaging demonstrated that the rate of change of signal enhancement is an important criteria for differentiating fibroadenomas from carcinomas (11,19). In most dynamic studies, recording parameters have been optimized to increase temporal resolution at the expense of spatial resolution (20–24). Analyses based on pharmacokinetic models were performed in selected regions of interest or over the whole tumor, resulting in loss of detailed spatial information (20–22,25–31). An overlap in the dynamic enhancement profiles of fibroadenoma and carcinoma indicates that the dynamic enhancement alone may not be sufficient for differentiating between these lesions (17). Another approach, on the basis of an interpretation model that incorporates breast MR architectural features, was recently presented that has high sensitivity and improved specificity for diagnosis of breast cancer (32–35).

Recently, we developed a method of contrast-enhanced MR imaging with which high-spatial-resolution images are obtained at three time points (one precontrast and two postcontrast time points) and then analyzed at the same high spatial resolution (1). In the final analysis, estimates are yielded of the microvascular permeability multiplied by the capillary surface area per pixel and the extracellular volume fraction. Herein, we present results in a pilot study of fibroadenomas that were diagnosed with conventional imaging and clinical criteria and confirmed with histopathologic findings. We tested the hypothesis that fibroadenomas possess unique values and distribution of two pathophysiologic features, microvascular permeability and extracellular volume fraction. The results demonstrated a correlation between imaging findings of the pathophysiologic tissue characteristics derived with the three-time-point method and the histopathologic findings.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Appendix
 References
 
MR imaging was performed in five consecutive female patients (age range, 19–60 years; mean age, 38.5 years) selected by a dedicated, experienced mammographer (S.S). On the basis of findings at clinical, mammographic, and US examinations, the lesions were consistent with fibroadenomas. Three patients had one lesion, one patient had two lesions in the right breast, and the remaining patient had one lesion in the left breast and two lesions in the right breast. All patients had been referred for surgery before they underwent MR imaging. Neither the diagnosis nor the referral for surgery was made on the basis of findings at MR imaging. The indications for removal of these benign tumors included (a) an enlarging palpable mass, (b) a nonpalpable enlarging mass, or (c) a nonpalpable de novo mass. The final diagnosis for the seven lesions was based on findings at histopathologic examination of the surgically removed lesions. One small lesion identical to and adjacent to a large lesion was not examined histopathologically. Informed consent was obtained from all patients before MR imaging and surgical biopsy.

MR Imaging
MR images were obtained with a spectrometer (2T Gyrex; Elscint, Haifa, Israel) supplied with a specially designed breast coil (36). A fast three-dimensional, gradient-echo sequence was used with a temporal resolution of 2–4 minutes (2 minutes for 32 sections, 4 minutes for 64 sections). The imaging parameters included the following: repetition time msec/echo time msec of 20/8.5, flip angle of 25°, field of view of 22 x 22 cm, and reconstruction matrix of 256 x 256 x 64 or 256 x 256 x 32. Gadopentetate dimeglumine was hand injected as a bolus at a dose 0.2 mmol per kilogram of body weight via a 20-gauge intravenous cannula prepositioned in an antecubital fossa vein and was followed by a 20-mL saline solution flush.

Histopathologic Findings
The lesions were surgically removed, fixed in 4% formaldehyde in saline solution, and embedded in paraffin. Serial 4-µm-thick sections were stained with hematoxylin-eosin and immunostained with anti–von Willebrand factor (factor VIII). Pathologic characteristics of all lesions were evaluated by inspecting the standard hematoxylin-eosin–stained slides, and microvessel distribution and density were assessed by examining slides immunostained for endothelial cells with factor VIII. The initial inspection of the histopathologic slices was performed independently by an expert pathologist (P.C.) without any prior knowledge of the MR imaging findings. This analysis included diagnosis of the lesion, characterization of the features of epithelial and stromal components, and qualitative assessment of the cellular volume fraction and the mucinous nature or hyalinization of the connective tissue element (Table 1). Variations in the vascularization in various regions of the lesions were also qualitatively estimated (Table 2).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Histopathologic Features of Breast Fibroadenomas
 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Vascular Features of Breast Fibroadenomas
 
Image Analysis
All images were loaded and analyzed on a workstation (DEC-3000; Digital Equipment, Herzelia, Israel). Analysis of the contrast enhancement in the tumors was performed at pixel resolution (800 x 800 x 1,200 mm) by employing a new program developed recently in our laboratory (1). The resultant images are color coded on the basis of hue and intensity as described in the Appendix (Fig 1). A separate program was developed to provide calibration maps based on a physiologic model for translating the color hue or intensity into estimates of the pathophysiologic parameters. A typical calibration map obtained for analyzing the three-time-point images of the fibroadenomas is shown in Figure 2.



View larger version (37K):
[in this window]
[in a new window]
  		Figure 1. Schematic representation of the color coding scheme of the three-time-point method. The three color hues display different patterns of contrast agent washout, and the intensities of each color display the initial wash-in rates.

 


View larger version (11K):
[in this window]
[in a new window]
  		Figure 2. Calibration map for referencing the color hue or color intensity of the fibroadenomas on the three-time-point images to the microvascular permeability and extracellular volume fraction. The map was constructed for the three time points t0 (precontrast) and t1 (4.5 minutes) and t2 (12 minutes) (postcontrast) as described in Materials and Methods, Image Analysis.

 
The experimental and calculated parameters used in the three-time-point analysis were the same for all patients. Gradient-echo MR imaging (20/8.5) was performed with a flip angle of 25°, spatial resolution of 800 x 800 x 1,200 µm, temporal resolution of 2 or 4 minutes (32 or 64 sections, respectively, depending on breast size). The vascular pharmacokinetic parameters (37) included amplitudes of a1 = 3.99 and a2 = 4.78 kg/L and rate constants of m1 = 0.144 and m2 = 0.011 per minute, nuclear relaxation time of tissue of T10 = 0.71 second (38), and relaxivity of r1 = 4.5 mmol-1 · L · sec-1 (39). The dose of gadopentetate dimeglumine was 0.2 mmol/kg.

The center of the acquisition time of each set of contrast-enhanced images was assigned as the postcontrast time point. For example, the 4-minute acquisition between 1.5 and 5.5 minutes after administration of contrast material was assigned as the 3.5-minute time point. The two postcontrast time points varied only slightly (±1 minute) for the different patients, and therefore each calibration map was very similar to that in Figure 2.

The MR images were processed with the three-time-point algorithm (D.W.), and the resultant three-time-point images were analyzed with prior knowledge of the diagnosis of the lesions as fibroadenomas. A final comparison study of the histopathologic slices and the three-time-point images was performed (D.W., S.S., P.C., S.F., and H.D.). Note that determination of the extracellular volume fraction (EVF), a parameter evaluated with the three-time-point method, also yields the cellularity (C), namely, the fraction of the tumor occupied by the cell volume (EVF + C = 1).


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Appendix
 References
 
The three-time-point method was applied to analyze lesions demonstrated at clinical examination, mammography, and US. The initial diagnosis was made before the MR imaging examination. Finally, the lesions were removed surgically and were reexamined histopathologically. We obtained histopathologic slides for seven of the eight lesions. Histopathologic examination was not performed in lesion 6 (Tables 1, 2), the small lesion close to a large fibroadenoma, and the diagnosis was presumed to be the same on the basis of the similar appearance of the two lesions on the three-time-point images. The histopathologic characterization of the features of epithelial and stromal components of the fibroadenomas and a qualitative assessment of the cellular volume fraction and of the mucinous nature or hyalinization of the connective tissue element have been summarized in Table 1. Qualitative estimates of the extent of vascularization in the various parts of the lesions are summarized in Table 2.

Figure 3 shows images of a fibroadenoma (lesion 8) of a mixed intra- and pericanalicular type in a 19-year-old patient. The contrast-enhanced MR images show another, much larger tumor in the same breast that will be described later. The three-time-point images of the former tumor exhibit mainly dark red pixels except in the anterior aspect of the surrounding capsule, which appears blue with some green pixels between the red and the blue pixels. According to the calibration map, the dark red pixels indicate tissue with low microvascular permeability and relatively high extracellular volume fraction, while the blue pixels indicate high microvascular permeability and a relatively lower extracellular volume fraction. The green pixels indicate intermediate values of these two parameters. The histopathologic slices were sectioned closely parallel to the imaging plane and were stained with hematoxylin-eosin (part A) and factor VIII (part B) to correlate with the three-time-point, color-coded images. The microscopic view of the endothelial cells (part C) indicates the presence of a homogeneous distribution of capillaries with medium and large blood vessels appearing only in the capsule.



View larger version (61K):
[in this window]
[in a new window]
  		Figure 3. Lesion 8 (a small, mixed fibroadenoma). The arrows on the t1 and t2 MR images indicate the location of the fibroadenoma depicted in the three-time-point images and in the histopathologic slices. The other large enhanced lesion is described in Figure 7. The three-time-point images demonstrate dark red pixels for most of the lesion, surrounded at the edges by blue pixels (arrows). In A–C, the histologic slices were stained with hematoxylin-eosin (A [original magnification, x16]) and with factor VIII (B [original magnification, x16] and C [original magnification, x100]), and the lesion was seen to be a vascularized capsule (arrows).

 
Figure 4 shows images of another fibroadenoma (lesion 4) with histopathologic slices obtained closely parallel to the imaging plane in a 20-year-old patient. The macroscopic and microscopic views of the histopathologic slices clearly demonstrate that the mass comprises several nodules and is of an intracanalicular type. Each nodule is surrounded by a capsule, which shows staining of endothelial cells with factor VIII (part B). The density of the blood vessels in the stroma of the tumor appears to be relatively high. The three-time-point images confirm the presence of the nodular structure and of vascularized capsules, namely, blue pixels surround each nodule. In this tumor, a large fraction of the pixels are green in accordance with a higher capillary density (part C) and high extracellular volume fraction. The innermost regions in each nodule appear red, which is indicative of the presence of high extracellular volume fraction but lower microvascular permeability.



View larger version (68K):
[in this window]
[in a new window]
  		Figure 4. Lesion 4 (a multinodular fibroadenoma). The relatively dense, large blood vessels in the capsule (arrows) that surrounds each nodule are depicted on the three-time-point images as blue pixels and in the histopathologic slices stained with hematoxylin-eosin (A [original magnification, x16]) or factor VIII (B [original magnification, x16] and C [original magnification, x100]).

 
In the other breast of this patient, two lesions (lesions 5 and 6) were demonstrated on the contrast-enhanced images. The inferior lesion was large and was diagnosed as fibroadenoma histopathologically. The superior small lesion was not examined histopathologically but was depicted on the three-time-point images as having the same pathophysiologic characteristics as the large fibroadenoma. The three-time-point images indicated the presence of mostly dark red pixels in accordance with high extracellular volume fraction and low microvascular permeability as was expected on the basis of the histopathologic findings in the large hyalinized fibroadenoma. In a few regions, predominantly at the periphery of the tumor, blue and green pixels were observed in agreement with the presence of vessels. Findings at the histopathologic examination indicated the presence of a multinodular, highly integrated fibroadenoma of mostly intracanalicular type with focally distributed parts of pericanalicular type. In this fibroadenoma, the intraductal stroma occupied a relatively high fraction of the lesion (40%) and exhibited a high extracellular volume fraction and hyalinized fibrous tissue indicative of a dormant or quiescent phase of its life cycle. The more cellular periductal stroma also contained focal myxoid and hyalinized connective tissue.

Figure 5 shows the large (volume, 15 cm3) fibroadenoma (lesion 2) in a 49-year-old patient. The three-time-point images exhibit mainly dark red pixels. A three-dimensional representation of this tumor from two different views is included. For simplicity, the three-dimensional images were obtained with use of color hue at one constant intensity. This representation emphasizes again the predominant presence of the red pixels as compared with the rarer blue pixels seen in fibroadenomas. Although in general fibroadenomas are solid masses, the histopathologic findings in this mixed fibroadenoma (intracanalicular and pericanalicular) indicate the presence of a relatively high extracellular volume fraction that may be mucinous, myxoid, or hyalinized and is accessible to the contrast agent. Two slides stained with factor VIII exhibit small blood vessels distributed throughout the parenchyma in two typical ways: as a region richer in microvessels (part A) or lower in microvessel density (part B). We suggest that the intensity of the red pixels, dark versus light, reflects low versus high microvessel density, respectively.



View larger version (84K):
[in this window]
[in a new window]
  		Figure 5. Lesion 2 (a large mixed fibroadenoma). The boxes describe three-dimensional representations of the two-dimensional three-time-point images. The histopathologic slices were stained with factor VIII and depict greater vascularity in A (original magnification, x100) than in B (original magnification, x100). This finding explains the differences in the color intensity among the red pixels.

 
Figure 6 shows another fibroadenoma (lesion 1) with three-time-point images dominated by dark red pixels. In the histopathologic section, the cellularity and density of the microvessels in this fibroadenoma are very low, in accord with the features deduced from the three-time-point images. The highly hyalinized features also indicate a dormant or quiescent state, which is in accord with the low vascularization. Unlike the borders of the other fibroadenomas in our study, which were smooth, the borders of this tumor were less well defined.



View larger version (79K):
[in this window]
[in a new window]
  		Figure 6. Lesion 1 (a fibroadenoma in an older patient). The arrows on the MR images indicate the location of the lesion. Among the three-time-point images, the square indicates the section shown in the MR images. The histopathologic slice (factor VIII stain; original magnification, x100) demonstrates the low vascularity and the high extracellular volume fraction, which are depicted as dark red in the three-time-point images.

 
Figure 7 shows a fibroadenoma (lesion 7) with very large multinodular septa in a 19-year-old patient. In this and some other fibroadenomas, a moderate washout, or green pattern was seen, and they were categorized as "young" lesions. The three-time-point images show clearly a vascularized capsule at the anterior part of the tumor (blue pixels) and well-defined septa that penetrate into the tumor from the outer capsule. These septa appear dark red with a parallel blue line, which is indicative of the presence of dense vascularity. Most of the pixels are coded green in accord with the presence of relatively high stromal vascularization, but pixels coded red, which is typical of regions with less permeable vessels, appear in the central part of each division. The three-time-point image on the right was obtained at the edge of the tumor and shows blue pixels in the surrounding capsule. Despite the large size of this mass, the three-time-point images are very similar and the color distribution is relatively homogeneous. In accordance with the three-time-point images, the hematoxylin-eosin–stained histopathologic slice (part A) showed large septa that cut through the center of the mass and divided it into three nodules. The extra-cellular volume fraction of these septa is high relative to that of the surrounding stroma, which is mainly of an intracanalicular type. A large blood vessel that crosses through the septum is depicted with red blood cells in its lumen (arrow). The histopathologic slice stained for endothelial cells (part B) demonstrates a relatively high density of small and medium blood vessels dispersed in the periductal stroma.



View larger version (81K):
[in this window]
[in a new window]
  		Figure 7. Lesion 7 (a large intracanalicular fibroadenoma). The arrows on the MR images indicate the location of the lesion. Among the three-time-point images, the square indicates the section shown in the MR images, and the fibroadenoma is depicted with three central sections and a section at the edge (on the left). In A (hematoxylin-eosin stain; original magnification, x100), the crossing area of the three septa is seen, as is a large blood vessel (arrow). In B (factor VIII stain; original magnification, x100), microvessels (arrows) are seen that correlate with the green pixels in the three-time-point images.

 
Lesion 3 was also depicted (figure not provided) with a large fraction of the red and green pattern, in accord with the lower extracellular volume fraction of its periductal fibrous stroma, which composed approximately 80% of the tumor, and with the presence of stromal blood vessels. On three-time-point images, its appearance was similar to that of lesion 4 (Fig 4) and of lesion 7 (Fig 7), in accord with its being a "young" fibroadenoma, exhibiting higher microvascular permeability indicated by the green color and its high intensity. In this relatively small, intracanalicular type of tumor, a few blue pixels appear in some of the images, primarily in the margins, indicative of capsular vasculature. In one of the images, blue pixels also appear in the center, in accord with findings in the histopathologic slice of a septum with blood vessels that penetrated into the tumor from the capsule.


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Appendix
 References
 
Our findings led us to conclude that our high-spatial-resolution MR imaging method was capable of depicting the gross histopathologic features characteristic of fibroadenomas. (a) The masses were quite homogeneous and contained a relatively high extracellular volume fraction that was accessible to the contrast agent. (b) The borders of the masses were well defined and usually smooth. (c) The architecture of most of the tumors consisted of a capsule that contained medium to large blood vessels in part of its circumference. (d) Septa, usually emerging from the capsule, contained predominantly medium blood vessels. (e) Small microvessels were dispersed quite evenly among the internal periductal and interductal stroma with variations in the density of vessels in different fibroadenomas.

In addition to these common features, we recognized two different groups of fibroadenomas. The first group was termed "old" on the basis of findings on the three-time-point images of mainly dark red pixels, which signify low microvascular permeability and high extracellular volume fraction. Histopathologically, the low vascular permeability was verified in most cases by means of factor VIII staining, which showed a lower density of small vessels in the periductal stroma (Table 2). In most cases, these fibroadenomas also exhibited a partially hyalinized stroma in accord with the presence of high extracellular volume fraction. The second group was termed "young" on the basis of findings on the three-time-point images of light red pixels and a higher proportion of green pixels, which is indicative of a higher microvascular permeability and a lower extracellular volume fraction. Histopathologically, the higher microvascular permeability was verified in most cases by means of factor VIII staining as a density of periductal small vessels higher than that of the first group (Table 2). The lower extracellular volume fraction was indicated by a higher cellularity and lack of aging processes (eg, hyalinization).

In all fibroadenomas, capsular and septal blood vessels appear to be depicted with the three-time-point method as blue pixels indicative of a high value for microvascular permeability multiplied by the surface area. Near the blue pixels are green pixels, which are predominantly associated with stroma with appreciable blood supply and high cellular volume fraction. The red pixels that appear next to the green pixels seem to be associated with the higher extracellular volume fraction of the inner stromal elements, with light and dark red pixels indicative of higher and lower microvascular permeability, respectively.

In previous MR imaging studies of fibroadenomas with histopathologic correlation, findings on T1-weighted images did not help distinguish normal fibroglandular tissue from fibroadenomas (33). On T2-weighted images, the contrast of fibroadenomas compared with normal surrounding tissue varied from high to low intensity (18,33). In the majority of cases, the intensity of fibroadenomas was increased after contrast agent administration but the enhancement also occurred in carcinomas and was therefore not specific for fibroadenomas. The degree of enhancement in fibroadenomas was suggested to be different from that in carcinomas (17,19,29,40–42), but the overlap was too great and limited the ability to help distinguish between the two masses. On contrast-enhanced MR images, the shape, border characteristics, lobulation, and internal septations appear to provide characteristic features that help distinguish fibroadenomas from carcinomas (18). The aging of fibroadenomas, which is associated with sclerosis and calcification, was demonstrated with lowered intensity and the absence of contrast enhancement on T2-weighted images (18). It appears that the pixel-by-pixel analysis with three-time-point imaging yields a better approximation of the histopathologic appearance of a fibroadenoma in terms of vascularity and cell volume fraction. Thus, mapping of these two features may improve the specificity of diagnosis of breast disease with high-spatial-resolution, contrast-enhanced MR imaging.

In summary, we correlated findings with three-time-point, contrast-enhanced MR imaging with histopathologic findings and tumor angiogenesis. Visually, microvascular permeability and extracellular volume fraction of fibroadenomas correlated well between the histopathologic slices and the three-time-point images. We performed this pilot study in anticipation of a prospective, full clinical trial with blinded readings in a population referred for biopsy because of high mammographic suspicion. The capability of this method to reveal pathophysiologic features is probably not limited to fibroadenomas and could be extended to other lesions in the breast and in other organs.


   Appendix
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Appendix
References
 
The behavior of a contrast agent after the initial distribution in the vascular volume can be treated as the sum of wash-in and washout processes. These processes involve reversible permeation through the capillaries and diffusion from the intravascular compartment to the extravascular or extracellular compartments and the extracellular milieu. The kinetics of these processes can be modeled with equations that take into account the overall pharmacokinetics in the vascular system, as well as the local density and permeability to the contrast agent of the lesion vasculature and the extracellular volume fraction accessible to the contrast agent. The relative MR signal intensity is measured at three consecutive time points along the enhancement curve—precontrast (t0) and postcontrast (t1, t2)—to estimate the wash-in rates and washout patterns.

Washout patterns are estimated from the relative values of the two postcontrast intensities It1 and It2 according to one of three patterns, each coded with a specific color (Figure 1). (a) A slow washout pattern is defined as It1 < It2 and assigned the color red. (b) A moderate washout pattern is defined as It1 = It2 and assigned the color green. (c) A fast washout pattern is defined as It1 > It2 and assigned the color blue.

Wash-in rates are estimated on the basis of the increase in intensity between the precontrast and the first postcontrast time point to determine an initial rate: (It1 - It0)/(t1 - t0). This initial rate is coded for each washout pattern on the basis of color intensity (Fig 1). On the three-time-point images in this article, color signal intensities ranged between 0 and 255, with the fastest initial rate scaled to 255. Thus, with the three-time-point method, the washout pattern and wash-in rate are determined for each pixel of a lesion and are presented as a color hue and a color intensity, respectively.

Findings on the three-time-point images were interpreted pathophysiologically by constructing calibration maps to relate the wash-in rate and washout pattern to two parameters: the microvascular permeability multiplied by the surface area (K) and the extracellular volume fraction (EVF) (v1) (1). On the basis of the modeling approach of Tofts and Kermode (34), the relative enhancement of intensity (It1/It0) due to the presence of a contrast agent, for a gradient-echo sequence is

where r1 is 4.5 mmol-1 · L · sec-1, T10 is 0.71 seconds, repetition time (TR) is 20 msec, and echo time is 8.5 msec. The concentration of the contrast agent, Ctt, is determined by the dose of the contrast agent (D); the pharmacokinetic parameters, include amplitude (a1 = 3.99 kg/L, a2 = 4.78 kg/L) and rates (m1 = 0.144 min-1, m2 = 0.011 min-1); the microvascular permeability K; and the extracellular volume fraction v1 according to

where b1, b2, and b3 are the coefficients of the three exponents (2). Figure 2 demonstrates a typical calibration map obtained with these equations and parameters. This map was used in the interpretation of findings on the three-time-point images of the fibroadenomas.


   Acknowledgments
 
We thank Aharon Weinstein, MS, for helping with the three-dimensional representation of the images, and Edna Furman-Haran, MD, and Dov Grobgeld, MS, for helping develop the three-time-point method.


   Footnotes
 
Current address: Department of Surgery (B), Assaf Harofeh Medical Center, Zrifin, Israel. Back

Current address: Department of Surgery (B), Assaf Harofeh Medical Center, Zrifin, Israel. Back

Current address: The Rachel Nash Jerusalem Comprehensive Breast Clinic, Jerusalem, Israel. Back

Supported in part by the Canadian Women for the Weizmann Institute. D.W. was supported by Yeda Research and Development.

Address reprint requests to H.D.

Author contributions: Guarantor of integrity of entire study, H.D.; study concepts, D.W., H.D.; study design, S.S., S.F., J.M.G., H.D.; definition of intellectual content, D.W., H.D.; literature research, D.W., H.D.; clinical studies, S.S., P.C., S.F., J.M.G.; experimental studies, D.W., S.S., P.C., S.F.; data acquisition, S.S., S.F., J.M.G., P.C., D.W.; data analysis, D.W., P.C., S.S., S.F., H.D.; manuscript preparation, D.W., H.D.; manuscript editing, S.S., S.F., J.M.G.; manuscript review, S.S., P.C., S.F., J.M.G.

Received February 24, 1998; revision requested April 22, 1998; revision received June 16, 1998; accepted August 20, 1998.
   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Appendix
 References
 

  1. Degani H, Gusis V, Weinstein D, Fields S, Strano S. Mapping pathophysiological features of breast tumors by MRI at high spatial resolution. Nature Med 1997; 3:780-782.[Medline]
  2. Turkat TJ, Klein BD, Polan RL, Richman RH. Dynamic MR mammography: a technique for potentially reducing the biopsy rate for benign breast disease. JMRI 1994; 4:563-568.
  3. Flinkiger FW, Allison JK, Sherry RM, Wright JC. Differentiation of benign from malignant breast masses by time-intensity valuation of contrast enhanced MRI. J Magn Reson Med 1993; 11:617-620.
  4. Fobben ES, Rubin CZ, Kalisher L, Dembner AG, Seltzer MH, Santoro EJ. Breast MR imaging with commercially available techniques: radiologic-pathologic correlation. Radiology 1995; 196:143-152.[Abstract/Free Full Text]
  5. Cortan RS, Kumar V, Robbins SL. Robbins pathologic basis of disease 4th ed. Edinburgh, UK: Livingstone, 1989; 1189-1190.
  6. Ahmed A. Diagnostic breast pathology: a text and colour atlas Edinburgh, UK: Livingstone, 1992; 72-76.
  7. Bland KI, Copeland EM. The breast: comprehensive management of benign and malignant diseases Philadelphia, Pa: Saunders, 1991.
  8. Obermair A, Czerwenka K, Kurz C, Kaider A, Sevelda P. Tumoral vascular density in breast tumors and their effect on recurrence-free survival. Chirurg 1994; 65:611-615.[Medline]
  9. Holcombe C, Pugh N, Lyons K, Douglas-Jones A, Mansel RE, Horgan K. Blood flow in breast cancer and fibroadenoma estimated by color Doppler ultrasonography. Br J Surg 1995; 82:787-788.[Medline]
  10. Rigauts H, Casselman J, Steyaert L, Devlies F, Pattyn G. Contribution of MRI and color Doppler sonography in breast cancer diagnosis. J Belge Radiol 1993; 76:226-231.[Medline]
  11. Kaiser WA. Dynamic magnetic resonance breast imaging using a double breast coil: a step toward routine examination of the breast. Front Eur Radiol 1990; 7:39-68.
  12. Harms SE, Flamig DP. MR imaging of the breast. JMRI 1993; 3:277-283.
  13. Stomper PC, Herman S, Klippenstein DL, et al. Suspect breast lesions: findings at dynamic gadolinium-enhanced MR imaging correlated with mammographic and pathologic features. Radiology 1995; 197:387-395.[Abstract/Free Full Text]
  14. Heywang-Kobrunner SH, Haustein J, Paul C, et al. Contrast-enhanced MR imaging of the breast: comparison of two different doses of gadopentetate dimeglumine. Radiology 1994; 191:639-646.[Abstract/Free Full Text]
  15. Terao E, Takeuchi H, Iwamura A, Murakami Y, Harada J, Tada S. Ability of subtraction and dynamic MR imaging to detect breast tumors: comparison with ultrasonography and mammography. Nippon Igaku Hoshasen Gakkai Zasshi 1994; 54:950-957.[Medline]
  16. Gribbestad IS, Nilsen G, Fjosne HE, Kvinnsland S, Haugen OA, Rinck PA. Comparative signal intensity measurements in dynamic gadolinium-enhanced MR mammography. JMRI 1994; 4:477-480.
  17. Orel SG, Schnall MD, Livolsi VA, Troupin RH. Suspicious breast lesions: MR imaging with radiologic-pathologic correlation. Radiology 1994; 190:485-493.[Abstract/Free Full Text]
  18. Hochman MG, Orel SG, Powell CM, et al. Fibroadenomas: MR imaging appearances with radiologic-histopathologic correlation. Radiology 1997; 204:123-129.[Abstract/Free Full Text]
  19. Kaiser WA, Zeitler E. MR imaging of the breast: fast imaging sequences with and without Gd-DTPA. Radiology 1989; 170:681-686.[Abstract/Free Full Text]
  20. Hoffmann U, Brix G, Knopp MV, Heb T, Lorenz WJ. Pharmacokinetic mapping of the breast: a new method for dynamic MR mammography. Magn Reson Med 1995; 33:506-514.[Medline]
  21. Frouge C, Guinebretiere JM, Contesso G, Paola RD, Blery M. Correlation between contrast enhancement in dynamic magnetic resonance imaging of the breast and tumor angiogenesis. Invest Radiol 1994; 12:1043-1049.
  22. Hulka CA, Smith BL, Sgroi DC, et al. Benign and malignant breast lesions: differentiation with echo-planar MR imaging. Radiology 1995; 197:33-38.[Abstract/Free Full Text]
  23. Kennedy SD, Szczepaniak LS, Gibson SL, Hilf R, Foster TH, Bryant RG. Quantitative MRI of Gd-DTPA uptake in tumors: response to photodynamic therapy. Magn Reson Med 1994; 31:292-301.[Medline]
  24. Perman WH, Heiberg EM, Grunz J, Herrmann VM, Janney CG. A fast 3D-imaging technique for performing dynamic Gd-enhanced MRI of breast lesions. Magn Reson Imaging 1994; 12:545-551.[Medline]
  25. Tofts PS, Berkowitz B, Schnall MD. Quantitative analysis of dynamic Gd-DTPA enhancement in breast tumors using a permeability model. Magn Reson Med 1995; 33:564-568.[Medline]
  26. Knopp MV, Heb T, Hoffmann U, et al. Assessment of different quantification techniques for MR-mammography (abstr) In: Proceedings of the Third Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1995; 1599.
  27. Hylton N, Frankel SD, Esserman LJ, Moore K, Sickles E. High resolution 3D maps of contrast enhancement patterns in breast tumors (abstr) In: Proceedings of the Third Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1995; 439.
  28. Buckley DL, Mussurakis S, Blackband SJ, Drew PJ, Horsman A. Dynamic MR mammography: a comparison of two pharmacokinetic modeling techniques (abstr) In: Proceedings of the Third Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1995.
  29. Kelcz F, Santyr GE, Cron GO. Application of a quantitative model to differentiate benign from malignant breast lesions detected by dynamic, Gd-enhanced MRI. Magn Reson Med 1996; 6:743-752.
  30. Mussurakis S, Buckley DL, Bowsley SJ, et al. Dynamic contrast-enhanced magnetic resonance imaging of the breast combined with pharmacokinetic analysis of gadolinium-DTPA uptake in the diagnosis of local recurrence of early stage breast carcinoma. Invest Radiol 1995; 30:650-662.[Medline]
  31. den Boer JA, Hoenderop RKKM, Smink J, et al. Pharmacokinetic analysis of Gd-DTPA enhancement in dynamic three dimensional MRI of breast lesions. J Magn Reson Med 1997; 7:702-715.
  32. Nunes LW, Schnall MD, Orel SG, et al. Breast MR imaging: interpretation model. Radiology 1997; 202:833-841.[Abstract/Free Full Text]
  33. Nunes LW, Scnall MD, Siegelman ES, et al. Diagnostic performance characteristics of architectural features revealed by high spatial resolution MR imaging of the breast. AJR 1997; 169:409-415.[Abstract/Free Full Text]
  34. Tofts PS, Kermode AG. Measurement of the blood-brain barrier permeability and leakage space using dynamic MR imaging. I. Fundamental concepts. Magn Reson Med 1991; 17:357-367.[Medline]
  35. Degani H, Fields S, Catane R, Gomori JM, Peretz T, Libove A. Evaluation of breast cancer therapy with contrast enhanced MRI at high spatial resolution. JMRI 1994; 4:116.
  36. Stelling CB, Powell DE, Mattingly SS. Fibroadenomas: histopathologic and MR imaging features. Radiology 1987; 162:399-407.[Abstract/Free Full Text]
  37. Weidmann HJ, Laniado M, Mutzel W. Pharmacokinetic of GdDTPA/dimeglumine after intravenous injection into healthy volunteers. Physiol Chem Phys Med NMR 1984; 16:167-172.[Medline]
  38. Poon CS, Bronskill MJ, Henkelman RM, Boyd NF. Quantitative magnetic resonance imaging parameters and their relationship to mammographic pattern. J Natl Cancer Inst 1992; 84:777-781.[Abstract/Free Full Text]
  39. Laufer RB. Paramagnetic metal complexes as water proton relaxation agents for NMR imaging: theory and design. Chem Rev 1987; 87:901-927.
  40. Stack JP, Redmond OM, Codd MB, Dervan PA, Ennis JT. Breast disease tissue characterization with Gd-DTPA enhancement profiles. Radiology 1990; 174:491-494.[Abstract/Free Full Text]
  41. Merchant TE, de Graf PW, Neiwenhuizen WEA, et al. Fibroadenoma of the breast: in vivo magnetic resonance characterization. Eur J Radiol 1991; 13:91-95.[Medline]
  42. Kelcz F, Santyr GE, Cron GO. Incorporation of wash-in and washout criteria for improvement of specificity in dynamic gadolinium-enhanced MRI of breast lesions (abstr) In: Proceedings of Third Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 1995; 435.



This article has been cited by other articles:


Home page
 RadiologyHome page
Q. Zhu, E. B. Cronin, A. A. Currier, H. S. Vine, M. Huang, N. Chen, and C. Xu
Benign versus Malignant Breast Masses: Optical Differentiation with US-guided Optical Imaging Reconstruction
Radiology, October 1, 2005; 237(1): 57 - 66.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
S. Meisamy, P. J. Bolan, E. H. Baker, M. G. Pollema, C. T. Le, F. Kelcz, M. C. Lechner, B. A. Luikens, R. A. Carlson, K. R. Brandt, et al.
Adding in Vivo Quantitative 1H MR Spectroscopy to Improve Diagnostic Accuracy of Breast MR Imaging: Preliminary Results of Observer Performance Study at 4.0 T
Radiology, August 1, 2005; 236(2): 465 - 475.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
K. Kitagawa, H. Sakuma, N. Ishida, T. Hirano, A. Ishihara, and K. Takeda
Contrast-Enhanced High-Resolution MRI of Invasive Breast Cancer: Correlation with Histopathologic Subtypes
Am. J. Roentgenol., December 1, 2004; 183(6): 1805 - 1809.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
M. A. Jacobs, P. B. Barker, D. A. Bluemke, C. Maranto, C. Arnold, E. H. Herskovits, and Z. Bhujwalla
Benign and Malignant Breast Lesions: Diagnosis with Multiparametric MR Imaging
Radiology, October 1, 2003; 229(1): 225 - 232.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 Weinstein, D.
Right arrow Articles by Degani, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Weinstein, D.
Right arrow Articles by Degani, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE