(Radiographics. 1999;19:79-92.)
© RSNA, 1999
Correlation of Lesion Appearance and Histologic Findings for the Nodes of a Breast MR Imaging Interpretation Model
Linda White Nunes, MD, MPH1,
Mitchell D. Schnall, MD, PhD2,
Susan G. Orel, MD2,
Mary G. Hochman, MD3,
Curtis P. Langlotz, MD, PhD2,
Carol A. Reynolds, MD2 and
Michael H. Torosian, MD2
1 Department of Radiology, Hahnemann University Hospital, Broad and Vine Sts, Philadelphia, PA 19102-1192 (L.W.N.)
2 Department of Radiology, University of Pennsylvania Medical Center, Philadelphia (M.D.S., S.G.O., C.P.L., C.A.R., M.H.T.)
3 Department of Radiology, Beth Israel Deaconess Medical Center, Boston, Mass (M.G.H.).
 |
Abstract
|
|---|
An interpretation model for evaluating magnetic resonance (MR) images of the breast was constructed that allowed differentiation of benign from malignant palpable or mammographically visible abnormalities. Architectural features define each node of the model. Investigation was subsequently made of the histologic findings in individuals within each node and of the frequency with which each histologic finding manifested as a particular architectural feature to determine whether nodal location and specific histologic findings are mutually predictive. The strongest associations were found between fibrocystic change and smooth masses, fibroadenoma and lobulated masses with nonenhancing internal septations, invasive ductal carcinoma (with or without ductal carcinoma in situ [DCIS]) and enhancing irregular or spiculated masses, invasive tubular carcinoma or radial scar and spiculated masses, medullary or colloid carcinoma and enhancing lobulated masses, invasive lobular carcinoma and the absence of a focal mass, DCIS and ductal enhancement, and DCIS (with or without invasive ductal carcinoma) and regional enhancement. Nodal location and histologic findings proved to be mutually predictive within the model; that is, the nodal location of MR imaging features within the model can be used to predict histologic findings and vice versa.
Index Terms: Breast, MR, 00.1214 • Images, interpretation • Magnetic resonance (MR), comparative studies, 00.1214
|
INTRODUCTION
|
|---|
The assessment of architectural features revealed at high-spatial-resolution magnetic resonance (MR) imaging of the breast facilitates the distinction of benign from malignant disease (1–9). Certain MR imaging findings are highly predictive of benign disease and include smooth or lobulated borders (negative predictive value [NPV] for malignancy = 97%–100%), absence of lesion enhancement (NPV = 100%), enhancement less than that of surrounding breast stroma (NPV = 93%–100%), and absence of a lesion (NPV = 92%) (2,10). The presence of nonenhancing internal septations in a smooth or lobulated lesion is highly specific for the diagnosis of fibroadenoma (specificity = 93%–97%) (2,4,5,11). In contrast, other findings are highly predictive of malignant disease, including spiculated borders (positive predictive value [PPV] for malignancy = 76%–88%) and rim enhancement (PPV = 79%–92%) (2,12,13). The use of contrast material allows better architectural assessment. However, investigators who compared the utility of architectural features (lesion conspicuity, signal intensity, contour, enhancement pattern) with that of contrast enhancement kinetics found the former to be superior to quantitative kinetic indices alone (P = .02) (3).
Prior to undertaking the present study, we developed an interpretation model for evaluating breast MR images (1) (Fig 1). This model allows differentiation of benign from malignant palpable or mammographically visible abnormalities and has the following diagnostic performance characteristics: sensitivity, 96%; specificity, 79%; PPV, 76%; NPV, 97%; and overall accuracy, 86%.
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Diagram shows a breast imaging interpretation model that incorporates data obtained in 192 patients. Each node of the tree-shaped model includes the defining architectural feature, the number of patients with cancer and benign disease, and the associated NPV and PPV. (Adapted and reprinted, with permission, from reference 1.)
|
|
In this article, we discuss and illustrate our investigation of histologic findings in the individuals within the nodes of that model. The investigation was undertaken to determine whether nodal location and specific histologic findings are mutually predictive. We illustrate the architectural features that define each node, detail the histologic findings in the individuals within each node, and summarize the frequency with which each histologic finding manifests as a given architectural feature.
|
MATERIALS AND METHODS
|
|---|
Having given their informed consent, 192 patients with mammographically visible or palpable lesions underwent MR imaging with a 1.5-T imager (Signa; GE Medical Systems, Milwaukee, Wis) and a surface-array coil designed for imaging the breast in gentle compression (4).
MR imaging evaluation included sagittal T1-weighted spin-echo imaging (repetition time msec/echo time msec = 500/15), fat-saturated T2-weighted fast spin-echo imaging (4,000/105), and contrast material–enhanced gradient-echo MR imaging with intravenous administration of 0.1 mmol/kg of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, New Jersey).
For most of the contrast-enhanced MR imaging, three-dimensional, dynamically enhanced, fat-saturated fast multiplanar spoiled gradient-echo images (25–28/4) were obtained with a 30° flip angle, a 16–18-cm field of view, a 512 x 256 matrix, and 2–3-mm sections with no gap. The images were obtained with a three-dimensional volume acquisition. Total imaging time was 7 minutes, and all images were obtained between 90 seconds and 3 minutes 40 seconds after injection of contrast material.
In the first 67 cases, contrast-enhanced images were obtained with a two-dimensional technique designed to study the time course of enhancement of the suspicious finding as well as the architectural features; thus, the entire breast was not visualized in a dynamically enhanced fashion. Fast multiplanar spoiled gradient-echo MR images (50/6–8, 45° flip angle) were obtained through the expected location of the finding as judged from T1- and T2-weighted images at 15–60-second intervals for 4–7 minutes. These images were obtained with an 8–14-cm field of view, a 256 x 128 matrix, and 3-mm-thick sections with a 1-mm gap. Rarely, a different T1-weighted fast multiplanar spoiled gradient-echo sequence was used inadvertently.
The radiologist supervising the MR examination had the patient's mammographic and physical examination results available to help localize the MR imaging region corresponding to the palpable or mammographically visible lesion. Any lesions discovered at MR imaging that were not seen at physical examination or mammography were excluded from the study.
All 192 patients subsequently underwent excisional biopsy (n = 186) or cyst aspiration (n = 6) for pathologic diagnosis of the breast lesions. Malignant lesions were classified as invasive ductal carcinoma; ductal carcinoma in situ (DCIS); invasive ductal carcinoma with DCIS; and nonductal carcinoma (invasive lobular, invasive tubular, medullary, and colloid carcinomas). Benign histologic findings included fibrocystic change; fibroadenoma; and other findings such as radial scar, lipoma, normal breast tissue, hyperplasia, and lobular carcinoma in situ. Invasive ductal carcinoma with DCIS was defined as invasive ductal carcinoma with any amount of DCIS present either within or outside the invasive component. Extensive intraductal component was defined as greater than 25% DCIS either within or outside the invasive component.
The MR imaging examinations of the first 98 patients were used to construct the model and were retrospectively reviewed concurrently by three radiologists (M.D.S., S.G.O., M.G.H.); those of the remaining 94 patients were used to validate the initial model and were prospectively evaluated concurrently by the same three radiologists. The radiologists were guided in their evaluation by a questionnaire listing a broad array of architectural features believed to have predictive value in distinguishing benign from malignant disease. All three radiologists were blinded to the pathologic findings, and at least two of the three were blinded to the mammographic and physical examination findings.
The model was constructed in a stepwise fashion, with each successive node occupied by the architectural feature that would isolate the largest number of cases with a high NPV for malignancy. After the model was tested prospectively in 94 cases, it was further expanded, and it is this expanded model that is discussed in this article.
Node-defining architectural features of the lesions included enhancement pattern (no discrete lesion; ductal, regional, focal); border (smooth, lobulated, irregular, spiculated); presence of nonenhancing internal septations; and degree of enhancement (none, minimal, moderate, marked). Minimal enhancement was defined as enhancement less than or equal to that of the surrounding fibroglandular breast tissue.
Statistics used to construct and validate the interpretation model are given in the original article (1). The positive and negative predictive values reported in the present article were derived from 2 x 2 contingency tables constructed to compare the stated histologic findings with the stated nodal locations and were analyzed with a two-tailed Fisher exact test.
|
RESULTS
|
|---|
Histologic Findings by Node
No Lesion.—In 38 patients, there was no abnormality visible at MR imaging that corresponded to the palpable or mammographically suspicious lesion. Thirty-five (92%) of these lesions were benign and three (8%) were malignant. (These and subsequent findings related to benignity or malignancy and the corresponding negative and positive predictive values are shown in Figure 1.) An NPV of 0.97 and a PPV of 0.03 are believed to better reflect this defining architectural feature (ie, no visible lesion) because two of the malignancies (invasive ductal carcinomas) were studied with the two-dimensional enhancement technique described earlier and were not located in the region of the breast chosen for dynamic enhancement. The third malignancy proved to be a 0.6 x 0.1-mm focus of DCIS that had manifested as a cluster of microcalcifications at mammography (see * in Fig 1). Histologic findings in the 38 patients included (in decreasing order of frequency) fibrocystic change (n = 22) (58%), other benign histologic findings (n = 10) (26%), fibroadenoma (n = 3) (8%), invasive ductal carcinoma (n = 2) (5%), and DCIS (n = 1) (3%). These and subsequent findings in the Results section are shown in the Table.
Ductal Enhancement.—Linear and branching foci of contrast enhancement thought to represent discrete ducts in DCIS were seen in 10 patients. Lesions were malignant in eight patients (80%) and benign in two patients (20%). Histologic findings included DCIS in four patients (40%) (Figs 2, 3), invasive ductal carcinoma in three patients (30%), fibrocystic change in two patients (20%), and invasive lobular carcinoma in one invasive ductal carcinoma and DCIS.
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Figures 2, 3. (2) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows ductal enhancement (arrow) in a 52-year-old woman with mammographically visible calcifications who proved to have DCIS. (Adapted and reprinted, with permission, from reference 1.) (3) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (26/4, 30° flip angle) shows ductal enhancement (arrow) in a 42-year-old woman with mammographically visible calcifications who proved to have DCIS.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Figures 2, 3. (2) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows ductal enhancement (arrow) in a 52-year-old woman with mammographically visible calcifications who proved to have DCIS. (Adapted and reprinted, with permission, from reference 1.) (3) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (26/4, 30° flip angle) shows ductal enhancement (arrow) in a 42-year-old woman with mammographically visible calcifications who proved to have DCIS.
|
|
Regional Enhancement.—In 19 patients, MR imaging demonstrated an ill-defined, diffuse pattern of enhancement not known to correspond to any specific benign or malignant abnormality. Lesions were malignant in 11 patients (58%) and benign in eight patients (42%). Histologic findings included fibrocystic change (n = 8) (42%) (Fig 4), DCIS (n = 4) (21%) (Fig 5), invasive ductal carcinoma with DCIS (n = 4) (21%), invasive lobular carcinoma (n = 2) (11%), and invasive ductal carcinoma (n = 1) (5%). Of the four patients with both invasive ductal carcinoma and DCIS, two had extensive intraductal component with over 25% DCIS present within the invasive component and all four had DCIS present outside the invasive component.
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Figures 4, 5. (4) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (28/4, 30° flip angle) shows multiple areas of regional enhancement (arrows) in a 39-year-old woman with a palpable mass who proved to have fibrocystic change. (5) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (28/4, 30° flip angle) shows regional enhancement (arrows) in a 49-year-old woman with mammographically visible calcifications who proved to have DCIS.
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Figures 4, 5. (4) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (28/4, 30° flip angle) shows multiple areas of regional enhancement (arrows) in a 39-year-old woman with a palpable mass who proved to have fibrocystic change. (5) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (28/4, 30° flip angle) shows regional enhancement (arrows) in a 49-year-old woman with mammographically visible calcifications who proved to have DCIS.
|
|
Focal Masses.—A discrete mass corresponding to a palpable or mammographically suspicious lesion was seen at MR imaging in 125 patients. Lesions were benign in 69 patients (55%) and malignant in 56 patients (45%). Histologic findings included fibrocystic change (n = 35) (28%), fibroadenoma (n = 27) (22%), invasive ductal carcinoma (n = 24) (19%), invasive ductal carcinoma with DCIS (n = 24) (19%), and other benign histologic findings (n = 7) (6%). In addition, there were two cases (2%) each of DCIS and invasive tubular carcinoma and one case (<1%) each of medullary carcinoma, colloid carcinoma (Fig 6), invasive lobular carcinoma, and combined invasive ductal and invasive lobular carcinoma.
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. T2-weighted fast spin-echo MR image (4,000/110) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/8, 45° flip angle) (b) show a focal mass (arrow) in a 75-year-old woman with a palpable mass that was visible at mammography and that proved to be colloid carcinoma.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. T2-weighted fast spin-echo MR image (4,000/110) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/8, 45° flip angle) (b) show a focal mass (arrow) in a 75-year-old woman with a palpable mass that was visible at mammography and that proved to be colloid carcinoma.
|
|
Focal masses with smooth borders were seen in 21 patients and were benign in all cases. Fifteen masses (71%) represented fibrocystic change and six (29%) represented fibroadenomas (Figs 7, 8).
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7. Figures 7, 8. (7) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows a smooth mass (arrow) in a 46-year-old woman who had calcifications and a mass that were visible at mammography and that proved to represent a fibroadenoma. The thick swirls of low signal intensity seen within the mass are a variant of the nonenhancing internal septations seen in association with fibroadenoma. (Adapted and reprinted, with permission, from reference 1.) (8) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (300/3, 90° flip angle) shows a very large smooth mass (arrows) in a 30-year-old woman with a palpable and mammographically visible mass that proved to be a giant fibroadenoma.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Figures 7, 8. (7) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows a smooth mass (arrow) in a 46-year-old woman who had calcifications and a mass that were visible at mammography and that proved to represent a fibroadenoma. The thick swirls of low signal intensity seen within the mass are a variant of the nonenhancing internal septations seen in association with fibroadenoma. (Adapted and reprinted, with permission, from reference 1.) (8) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (300/3, 90° flip angle) shows a very large smooth mass (arrows) in a 30-year-old woman with a palpable and mammographically visible mass that proved to be a giant fibroadenoma.
|
|
Focal masses with lobulated borders and nonenhancing internal septations were seen in 20 patients and were all benign. Histologic findings included fibroadenomas in 13 patients (65%) (Figs 9, 10), fibrocystic change in six patients (30%), and other benign histologic findings in one patient (5%).
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9. Figures 9, 10. (9) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/8, 45° flip angle) shows a lobulated mass (white arrow) with nonenhancing internal septations in a 21-year-old woman with a palpable mass that proved to be a fibroadenoma. Black arrow indicates an internal septation. (Adapted and reprinted, with permission, from reference 1.) (10) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 60° flip angle) shows a lobulated mass (white arrow) with nonenhancing internal septations in a 49-year-old woman with mammographically visible calcifications who proved to have a fibroadenoma. Black arrow indicates an internal septation.
|
|
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10. Figures 9, 10. (9) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/8, 45° flip angle) shows a lobulated mass (white arrow) with nonenhancing internal septations in a 21-year-old woman with a palpable mass that proved to be a fibroadenoma. Black arrow indicates an internal septation. (Adapted and reprinted, with permission, from reference 1.) (10) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 60° flip angle) shows a lobulated mass (white arrow) with nonenhancing internal septations in a 49-year-old woman with mammographically visible calcifications who proved to have a fibroadenoma. Black arrow indicates an internal septation.
|
|
Eleven patients had focal masses with lobulated borders, no nonenhancing internal septations, and minimal or no enhancement. All 11 masses were benign and were categorized at histologic analysis as fibroadenoma (n = 5) (45%), fibrocystic change (n = 3) (27%), or other benign histologic findings (n = 3) (27%). Figure 11 demonstrates a mature fibroadenoma. The low T2 signal reflects the sclerotic, fibrotic changes in a mature, involuting fibroadenoma (5).
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. T2-weighted fast spin-echo MR image (4,000/105) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/8, 60° flip angle) (b) show a minimally enhancing lobulated mass (arrows) in a 38-year-old woman with a mammographically visible mass that proved to be a fibroadenoma.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. T2-weighted fast spin-echo MR image (4,000/105) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/8, 60° flip angle) (b) show a minimally enhancing lobulated mass (arrows) in a 38-year-old woman with a mammographically visible mass that proved to be a fibroadenoma.
|
|
Fourteen patients had focal masses with lobulated borders, no nonenhancing internal septations, and moderate to marked enhancement. Eight masses (57%) were benign and six (43%) were malignant. Histologic findings included fibrocystic change in four patients (29%), fibroadenoma in three patients (21%), invasive ductal carcinoma in three patients (21%), and one case (7%) each of invasive ductal carcinoma with DCIS, medullary carcinoma (Fig 12), colloid carcinoma, and other benign histologic findings. The peripheral rim enhancement seen in Figure 12 and subsequent figures is strongly associated with malignancy (1,2,4,12).
View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12a. T2-weighted fast spin-echo MR image (4,000/100) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 45° flip angle) (b) show a rim-enhancing lobulated mass (arrows) in a 40-year-old woman with a palpable and mammographically visible mass that proved to be medullary carcinoma. Peripheral rim enhancement is strongly associated with malignancy.
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12b. T2-weighted fast spin-echo MR image (4,000/100) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 45° flip angle) (b) show a rim-enhancing lobulated mass (arrows) in a 40-year-old woman with a palpable and mammographically visible mass that proved to be medullary carcinoma. Peripheral rim enhancement is strongly associated with malignancy.
|
|
A focal mass with irregular borders and nonenhancing internal septations was seen in only one patient and was benign. The mass represented fibrocystic change (Fig 13).
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. T2-weighted fast spin-echo MR image (4,000/120) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 60° flip angle) (b) show an irregular mass (arrows) with nonenhancing internal septations in a 36-year-old woman with a palpable mass who proved to have fibrocystic change. The mass contains discrete cysts that have high T2 signal and do not enhance centrally with contrast material.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. T2-weighted fast spin-echo MR image (4,000/120) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 60° flip angle) (b) show an irregular mass (arrows) with nonenhancing internal septations in a 36-year-old woman with a palpable mass who proved to have fibrocystic change. The mass contains discrete cysts that have high T2 signal and do not enhance centrally with contrast material.
|
|
Twenty-six patients had focal masses with irregular borders and no nonenhancing internal septations. Twenty-two masses (85%) were malignant and four (15%) were benign. Histologic analysis revealed invasive ductal carcinoma with DCIS in 12 patients (46%) (Figs 14, 15), invasive ductal carcinoma in nine patients (35%), fibrocystic change in four patients (15%), and DCIS in one patient (4%).
View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14. Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows an irregular mass (arrow) in a 47-year-old woman with a palpable and mammographically visible mass that proved to be invasive ductal carcinoma with DCIS.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15. Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 45° flip angle) shows a rim-enhancing irregular mass (arrow) in a 48-year-old woman with a palpable mass that proved to be invasive ductal carcinoma with DCIS.
|
|
Focal masses with spiculated borders, no nonenhancing internal septations, and minimal or no enhancement were seen in two patients and were both benign. One mass represented radial scar with fibrocystic change (Fig 16), and the other represented fibrocystic change only.
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16a. T1-weighted spin-echo MR image (500/15) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 60° flip angle) (b) show a nonenhancing spiculated mass (arrow) in a 61-year-old woman with a mammographically visible mass who proved to have a radial scar and fibrocystic change.
|
|
View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16b. T1-weighted spin-echo MR image (500/15) (a) and contrast-enhanced fast multiplanar spoiled gradient-echo MR image (50/6, 60° flip angle) (b) show a nonenhancing spiculated mass (arrow) in a 61-year-old woman with a mammographically visible mass who proved to have a radial scar and fibrocystic change.
|
|
Focal masses with spiculated borders, no nonenhancing internal septations, and moderate or marked enhancement were seen in 30 patients. Twenty-eight (93%) of these masses were malignant and two (7%) were benign. There were 12 cases (40%) of invasive ductal carcinoma (Fig 17), 11 (37%) of invasive ductal carcinoma with DCIS, two (7%) of invasive tubular carcinoma (Fig 18), and one (3%) each of invasive lobular carcinoma, combined invasive ductal and invasive lobular carcinoma, DCIS, radial scar, and fibrocystic change.
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17. Figures 17, 18. (17) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows a rim-enhancing spiculated mass (arrow) in a 69-year-old woman with a mammographically visible mass that proved to be invasive ductal carcinoma. (18) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows a rim-enhancing spiculated mass (arrow) in a 64-year-old woman with mammographically visible calcifications who proved to have invasive tubular carcinoma.
|
|
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 18. Figures 17, 18. (17) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows a rim-enhancing spiculated mass (arrow) in a 69-year-old woman with a mammographically visible mass that proved to be invasive ductal carcinoma. (18) Contrast-enhanced fast multiplanar spoiled gradient-echo MR image (25/4, 30° flip angle) shows a rim-enhancing spiculated mass (arrow) in a 64-year-old woman with mammographically visible calcifications who proved to have invasive tubular carcinoma.
|
|
Nodal Distribution by Histologic Finding
Fibrocystic Change.—Nodal distribution of the 67 patients with fibrocystic change was as follows: 22 patients (33%) with no lesion visible at MR imaging, 15 (22%) with smooth masses, eight (12%) with regional enhancement, six (9%) with lobulated masses with septations, four (6%) with lobulated masses with no septations and moderate to marked enhancement, four (6%) with irregular masses with no septations, three (4%) with lobulated masses with no septations and minimal or no enhancement, two (3%) with ductal enhancement, one (1%) with an irregular mass with septations, one (1%) with a spiculated mass with minimal or no enhancement, and one (1%) with a spiculated mass with moderate to marked enhancement.
Fibroadenoma.—Nodal distribution of the 30 patients with fibroadenoma was as follows: 13 patients (43%) with lobulated masses with septations, six (20%) with smooth masses, five (17%) with lobulated masses with no septations and minimal or no enhancement, three (10%) with lobulated masses with no septations and moderate to marked enhancement, and three (10%) with no lesion visible at MR imaging.
Other Benign Histologic Findings.—Other benign histologic findings (n = 17) included two radial scars and one lipoma as well as normal breast tissue, hyperplasia, and lobular carcinoma in situ. The radial scars manifested as spiculated masses, one with minimal or no enhancement (6%) and the other with moderate to marked enhancement (6%). The lipoma manifested as a lobulated mass with no septations and minimal or no enhancement (6%). The remaining histologic findings had the following nodal distribution: 10 cases in which no lesion was visible at MR imaging (59%), two lobulated masses with no septations and minimal or no enhancement (12%), one lobulated mass with no septations and moderate to marked enhancement (6%), and one lobulated mass with septations (6%).
Invasive Ductal Carcinoma.—Invasive ductal carcinoma (n = 30) manifested as a spiculated mass with moderate to marked enhancement in 12 cases (40%), an irregular mass with no septations in nine cases (30%), a lobulated mass with no septations and moderate to marked enhancement in three cases (10%), ductal enhancement in three cases (10%), no visible lesion at MR imaging in two cases (7%), and regional enhancement in one case (3%). In both cases involving no visible lesion, no dynamic, gadolinium-enhanced sequence was used to image the involved region of the breast.
In the patient with combined invasive ductal carcinoma and invasive lobular carcinoma, the lesion manifested as a spiculated mass with moderate to marked enhancement (see * in Table).
DCIS.—In the 11 patients with DCIS, the lesions manifested as ductal enhancement (n = 4) (36%), regional enhancement (n = 4) (36%), a spiculated mass with moderate to marked enhancement (n = 1) (9%), and an irregular mass with no septations (n = 1) (9%). One case (9%) involved a 0.6 x 0.1-mm focus of DCIS that had manifested at mammography as a cluster of microcalcifications. This lesion was not visible at MR imaging.
Invasive Ductal Carcinoma with DCIS.—Invasive ductal carcinoma with DCIS was seen in 28 patients and manifested as irregular masses with no septations in 12 cases (43%), spiculated masses with moderate to marked enhancement in 11 cases (39%), regional enhancement in four cases (14%), and a lobulated mass with no septations and moderate to marked enhancement in one case (4%).
Nonductal Carcinoma.—Nonductal carcinoma was seen in nine patients and represented invasive lobular carcinoma in four cases; invasive tubular carcinoma in two cases; and medullary carcinoma, colloid carcinoma, and combined invasive ductal and invasive lobular carcinoma in one case each. The invasive lobular carcinomas manifested as regional enhancement in two cases (22%), as a spiculated mass with moderate to marked enhancement in one case (11%), and as ductal enhancement in one case (11%). The invasive tubular carcinomas manifested as spiculated masses with moderate to marked enhancement (22%). The medullary carcinoma appeared as a lobulated mass with no septations and moderate to marked enhancement (11%), as did the colloid carcinoma (11%).
|
DISCUSSION
|
|---|
When no lesion was seen at MR imaging, the pathologic diagnosis was almost always benign (NPV for malignancy 
92% [35 of 38], P < .01). The most common histologic finding was fibrocystic change (PPV = 58% [22 of 38], P < .01).
Smooth masses, lobulated masses with nonenhancing septations, and nonenhancing lobulated masses were always benign (NPV for malignancy = 100% [52 of 52], P < .01) and predominantly represented fibroadenoma (PPV = 46% [24 of 52], P < .01) or fibrocystic change (PPV = 46% [24 of 52], P < .01). Smooth masses most often represented fibrocystic change (PPV = 71% [15 of 21], P < .01), and lobulated masses with nonenhancing internal septations most often represented fibroadenoma (PPV = 65% [13 of 20], P < .01).
Other nonenhancing masses and other masses with nonenhancing septations were rare and were always benign. There was one nonenhancing irregular mass with nonenhancing septations that represented fibrocystic change and two nonenhancing spiculated masses that represented fibrocystic change or radial scar.
Enhancing irregular masses and enhancing spiculated masses without septations were almost always malignant (PPV for malignancy = 89% [50 of 56], P < .01) and usually represented invasive ductal carcinoma (PPV = 38% [21 of 56], P < .01) or invasive ductal carcinoma with DCIS (PPV = 41% [23 of 56], P < .01).
The most common histologic finding that manifested as ductal enhancement was DCIS (PPV = 40% [four of 10], P < .01). Regional enhancement was evenly split between fibrocystic change (PPV = 42% [eight of 19], P < .01) and malignancies with a DCIS component (PPV = 42% [eight of 19], P < .01). Of these malignancies, half represented DCIS and half represented invasive ductal carcinoma with DCIS.
Conversely, fibrocystic change most commonly manifested with no lesion visible at MR imaging (PPV = 33% [22 of 67], P < .01) or as a smooth mass (PPV = 22% [15 of 67], P < .01) but had a wide variety of MR imaging appearances. Fibroadenoma most commonly manifested as a lobulated mass with nonenhancing internal septations (PPV = 43% [13 of 30], P < .01) or a lobulated mass without septations (PPV = 27% [eight of 30], P < .01). Invasive ductal carcinoma, regardless of whether it was associated with DCIS, usually manifested as a spiculated mass (PPV = 40% [23 of 58], P < .01) or an irregular mass (PPV = 36% [21 of 58], P < .01). DCIS most commonly manifested as either ductal enhancement (PPV = 36% [four of 11], P < .01) or regional enhancement (PPV = 36% [four of 11], P < .01). Medullary carcinoma (n = 1) and colloid carcinoma (n = 1) were rare and manifested as enhancing lobulated masses. Invasive tubular carcinoma (n = 2) and radial scars (n = 2) were also rare and appeared as spiculated masses. Invasive lobular carcinoma did not usually manifest as a focal mass (NPV = 75% [three of four], P < .01), but as regional enhancement (two of four) or ductal enhancement (one of four).
|
CONCLUSION
|
|---|
Within our breast imaging interpretation model, nodal location and histologic findings are mutually predictive; that is, the nodal location of a patient within the model can be used to predict histologic findings and vice versa.
|
Acknowledgments
|
|---|
We thank Larry Muenz, PhD, for his statistical assistance; Jean McDermott, RN, for her assistance with patient care and database management; and Andrea Kaldrovics and Selicia Russo for their photographic contributions.
|
Footnotes
|
|---|
Supported by NIH grants R01-CA58358 and 5-T32-HL07614. L.W.N. supported in part by E-Z-Em.
Address reprint requests to L.W.N.
Presented as a scientific exhibit at the 1997 RSNA scientific assembly.
Abbreviations: DCIS = ductal carcinoma in situ
NPV = negative predictive value
PPV = positive predictive value
Received for publication March 13, 1998.
Revision received June 11, 1998. August 17, 1998.
Accepted for publication August 18, 1998.
|
References
|
|---|
-
Nunes LW, Schnall MD, Orel SG, et al. Breast MR imaging: interpretation model. Radiology 1997; 202:833-841.[Abstract/Free Full Text]
-
Nunes LW, Schnall MD, Siegelman E, 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]
-
Mussurakis S, Buckley DL, Drew PJ, et al. Dynamic MR imaging of the breast combined with analysis of contrast agent kinetics in the differentiation of primary breast tumours. Clin Radiol 1997; 52:516-526.[Medline]
-
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]
-
Hochman MG, Orel SG, Powell CM, Schnall MD, Reynolds CA, White L. Fibroadenomas: variety of MR imaging appearances with radiologic-histopathologic correlation. Radiology 1997; 204:123-129.[Abstract/Free Full Text]
-
Orel SG, Mendonca MH, Reynolds C, Schnall MD, Solin LJ, Sullivan DC. MR imaging of ductal carcinoma in situ. Radiology 1997; 202:413-420.[Abstract/Free Full Text]
-
Sherif H, Mahfouz AE, Oellinger H, et al. Peripheral washout sign on contrast-enhanced MR images of the breast. Radiology 1997; 205:209-213.[Abstract/Free Full Text]
-
Friedman EP, Hall-Craggs MA, Mumtaz H, Schneidau A. Breast MR and the appearance of the normal and abnormal nipple. Clin Radiol 1997; 52:854-861.[Medline]
-
Sinha S, Lucas-Quesada FA, DeBruhl ND, et al. Multifeature analysis of GD-enhanced MR images of breast lesions. JMRI 1997; 7:1016-1026.
-
Kelcz F, Santyr GE, Cron GO, Mongin SJ. Application of a quantitative model to differentiate benign from malignant breast lesions detected by dynamic, gadolinium-enhanced MRI. JMRI 1996; 6:743-752.
-
Soderstrom CE, Harms SE, Copit DS, et al. Three-dimensional RODEO breast MR imaging of lesions containing ductal carcinoma in situ. Radiology 1996; 201:427-432.[Abstract/Free Full Text]
-
Rubens D, Trotterman S, Chacko AK, et al. Gadopentetate dimeglumine–enhanced chemical-shift MR imaging of the breast. AJR 1991; 157:267-270.[Abstract/Free Full Text]
-
Buadu LD, Murakami J, Murayama S, et al. Patterns of peripheral enhancement in breast masses: correlation of findings on contrast medium enhanced MRI with histologic features and tumor angiogenesis. J Comput Assist Tomogr 1997; 21:421-430.[Medline]
This article has been cited by other articles:
|
|
|
|
 
B. Erguvan-Dogan, G. J. Whitman, A. C. Kushwaha, M. J. Phelps, and P. J. Dempsey
BI-RADS-MRI: a primer.
Am. J. Roentgenol.,
August 1, 2006;
187(2):
W152 - W160.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. Bartella, L. Liberman, E. A. Morris, and D. D. Dershaw
Nonpalpable mammographically occult invasive breast cancers detected by MRI.
Am. J. Roentgenol.,
March 1, 2006;
186(3):
865 - 870.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. K. Kuhl, H. H. Schild, and N. Morakkabati
Dynamic Bilateral Contrast-enhanced MR Imaging of the Breast: Trade-off between Spatial and Temporal Resolution
Radiology,
September 1, 2005;
236(3):
789 - 800.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. A. Berg, L. Gutierrez, M. S. NessAiver, W. B. Carter, M. Bhargavan, R. S. Lewis, and O. B. Ioffe
Diagnostic Accuracy of Mammography, Clinical Examination, US, and MR Imaging in Preoperative Assessment of Breast Cancer
Radiology,
December 1, 2004;
233(3):
830 - 849.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. Warner, D. B. Plewes, K. A. Hill, P. A. Causer, J. T. Zubovits, R. A. Jong, M. R. Cutrara, G. DeBoer, M. J. Yaffe, S. J. Messner, et al.
Surveillance of BRCA1 and BRCA2 Mutation Carriers With Magnetic Resonance Imaging, Ultrasound, Mammography, and Clinical Breast Examination
JAMA,
September 15, 2004;
292(11):
1317 - 1325.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. Huang, P. R. Fisher, K. Dulaimy, L. A. Tudorica, B. O'Hea, and T. M. Button
Detection of Breast Malignancy: Diagnostic MR Protocol for Improved Specificity
Radiology,
August 1, 2004;
232(2):
585 - 591.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Inoue, T. Sano, R. Watai, R. Ashikaga, K. Ueda, M. Watatani, and Y. Nishimura
Dynamic Multidetector CT of Breast Tumors: Diagnostic Features and Comparison with Conventional Techniques
Am. J. Roentgenol.,
September 1, 2003;
181(3):
679 - 686.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. Liberman, E. A. Morris, M. J.-Y. Lee, J. B. Kaplan, L. R. LaTrenta, J. H. Menell, A. F. Abramson, S. M. Dashnaw, D. J. Ballon, and D. D. Dershaw
Breast Lesions Detected on MR Imaging: Features and Positive Predictive Value
Am. J. Roentgenol.,
July 1, 2002;
179(1):
171 - 178.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. A. Morris, L. Liberman, D. D. Dershaw, J. B. Kaplan, L. R. LaTrenta, A. F. Abramson, and D. J. Ballon
Preoperative MR Imaging--Guided Needle Localization of Breast Lesions
Am. J. Roentgenol.,
May 1, 2002;
178(5):
1211 - 1220.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. W. Nunes, M. D. Schnall, and S. G. Orel
Update of Breast MR Imaging Architectural Interpretation Model
Radiology,
May 1, 2001;
219(2):
484 - 494.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
K. Kinkel, T. H. Helbich, L. J. Esserman, J. Barclay, E. H. Schwerin, E. A. Sickles, and N. M. Hylton
Dynamic High-Spatial-Resolution MR Imaging of Suspicious Breast Lesions: Diagnostic Criteria and Interobserver Variability
Am. J. Roentgenol.,
July 1, 2000;
175(1):
35 - 43.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
INTRODUCTION
