10.1245/ASO.2005.03.520
Annals of Surgical Oncology 12:1045-1053 (2005)
© 2005 Society of Surgical Oncology
Utility of Breast Magnetic Resonance Imaging in Patients With Occult Primary Breast Cancer
Claire L. Buchanan, MD1,
Elizabeth A. Morris, MD2,
Paige L. Dorn, BS1,
Patrick I. Borgen, MD1 and
Kimberly J. Van Zee, MD1
1 Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021
2 Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021
Correspondence: Address correspondence and reprint requests to: Kimberly J. Van Zee, MD; E-mail: vanzeek{at}mskcc.org
 |
ABSTRACT
|
|---|
Background: Although carcinoma presenting as axillary metastases is assumed to be due to breast cancer, identification of the primary lesion may prove problematic. We investigated the ability of breast magnetic resonance imaging (MRI) to identify the primary tumor, thereby confirming the diagnosis and broadening treatment options.
Methods: From 1995 to 2001, 69 patients at our institution presented with occult primary breast cancer. All patients had negative breast examinations and mammograms and underwent breast MRI.
Results: Of 69 patients, 55 had axillary adenopathy without evidence of distant disease (stage II); 14 had stage IV disease. In patients with stage II disease, MRI revealed suspicious lesions in 76% (42 of 55). In 62% (26 of 42), the MRI finding proved to be the occult primary tumor. Of these, 58% (15 of 26) were candidates for breast conservation. MRI did not identify the primary tumor in 25 women; 12 underwent mastectomy. Cancer was found in 33% (4 of 12) of these. Thirteen patients were treated with primary breast irradiation; three were lost to follow-up, one developed distant disease, and nine were without evidence of disease with a median follow-up of 4.5 years. In women with stage IV disease, MRI identified the primary tumor in 5 of 9 patients with regional adenopathy and 2 of 5 patients with distant disease (overall 50%; 7 of 14). MRI identified the primary tumor in women with both mammographically dense (19 of 44; 43%) and less dense (10 of 20; 50%) breasts.
Conclusions: Breast MRI detects mammographically occult cancer in half of women with axillary metastases, regardless of breast density. MRI is a powerful tool for stage II and stage IV patients with occult primary breast cancer.
Key Words: Occult primary breast carcinoma • Axillary metastases • Magnetic resonance imaging • Breast cancer • Breast imaging • Breast irradiation
|
INTRODUCTION
|
|---|
Magnetic resonance imaging (MRI) of the breast has been shown in previous studies to be a useful imaging tool for the management of breast cancer. In particular, it has been demonstrated to be beneficial as a screening tool in high-risk populations or those with a genetic predisposition.1–5 In addition, breast MRI can detect occult disease in the breast that is not appreciated on mammography or physical examination.4,6–9
Patients without evidence of breast disease who present with axillary lymph node metastases consistent with breast origin (TXN1M0) pose a management problem. Similarly challenging for the clinician are patients who present with distant metastases suggestive of a breast primary tumor (TXN0M1) when no primary tumor can be found. After initial investigations of physical examination and mammography fail to identify a primary breast tumor, many patients are advised to undergo a mastectomy. Up to a third of these patients will not have carcinoma identified at mastectomy with serial sectioning.10 Although the primary tumor may be so small that it evades pathologic detection, the lack of primary site identification raises the possibility that the site of origin is elsewhere. Another treatment option for these patients is whole-breast irradiation without surgery. However, this has been associated with a higher recurrence rate when compared with mastectomy, presumably as a result of inadequately treated disease.11
The advent of breast MRI now may give clinicians the ability to detect these previously occult lesions and potentially offer breast conservation. MRI technology has improved over the last decade, with better image resolution and improved biopsy capability. More significantly, increasing experience with breast MRI has resulted in more skilled interpretation of the images.
In a small series of 40 patients with biopsy-proven metastatic adenocarcinoma to an axillary lymph node from Memorial Sloan-Kettering Cancer Center (MSKCC), Olson et al.6 previously demonstrated that MRI of the breast identified a primary breast lesion in 70% of patients and that 47% of patients were able to preserve their breasts. Now that the technology of breast MRI has matured, we took this opportunity to reexamine our experience. In a larger population of patients, we sought to examine the utility of breast MRI in the setting of patients with occult primary breast carcinoma. Our criteria for inclusion differed from those of the prior series and included patients with regional adenopathy or distant disease consistent with breast origin. Our series included only patients with truly occult primary breast cancer (i.e., those without findings on physical examination or mammography) after thorough evaluation at MSKCC. A detailed review follows of how breast MRI affected the management of this unique patient group.
|
MATERIALS AND METHODS
|
|---|
After permission was obtained from our institutional review board, a retrospective review was conducted of all patients (n = 109) identified in a prospectively collected database as presenting with an occult primary breast cancer. These women were referred to the Breast Surgery Service at MSKCC from 1995 to 2001.
Mammography was performed at our institution or, if performed at an outside facility, was reviewed by a dedicated breast radiologist and repeated if the examination was not of adequate quality. All physical examinations were performed by a breast surgeon at our institution. Only patients with negative mammograms and completely negative breast examinations were considered to have truly occult disease and deemed suitable for inclusion in our analysis.
Each patient underwent a breast MRI at our institution, whose technique and interpretation has been described previously in detail.8 In 1998, MRI-directed localization capability became available. MRI-detected lesions that warranted biopsy included masses that had irregular or spiculated margins or lesions with heterogenous, linear, or segmental enhancement. Correlative sonography was recommended at the discretion of the radiologist interpreting the MRI examination if it was thought that the lesion might be sonographically evident and amenable to sonographically guided biopsy. If the lesion was not seen on sonography, MRI-guided needle localization for surgical excision was performed by using previously described methods.12
Of the 109 patients identified, 40 patients were excluded (reasons for exclusion are listed in Table 1
). Therefore, our patient population consisted of 69 patients who presented with biopsy-proven adenocarcinoma consistent with breast origin, who had a negative breast examination and negative mammogram, and who underwent a breast MRI at our institution.
Records of all 69 patients who underwent breast MRI for occult primary breast cancer were reviewed. Clinicopathologic data were recorded. A patient was considered to have a false-positive MRI if the biopsied lesion identified by the MRI examination did not reveal carcinoma. A patient was considered to have a false-negative MRI if there was no suspicious finding on MRI, yet the mastectomy specimen revealed carcinoma.
Status at last follow-up was obtained from clinic records. Mammographic density was recorded in the mammogram reports dictated by MSKCC dedicated breast radiologists. Breast density was scored according to the Breast Imaging Reporting and Data System13 classification, with scores of 1 through 4 as follows: 1, fatty; 2, mildly dense; 3, moderately dense; and 4, extremely dense.
|
RESULTS
|
|---|
The median age of our study population was 58 years (range, 23–80 years). Of patients who underwent complete axillary lymph node dissection, 49% (32 of 65) were found to be estrogen receptor positive, and 54% (35 of 65) were found to have nodal metastases in addition to the initial axillary metastasis. The 69 patients with occult primary breast cancer were divided into 2 groups: 55 stage II patients (TXN1M0) and 14 stage IV patients (TXN0M1). All of the stage II patients presented with isolated axillary adenopathy. Stage IV patients initially either presented with distant disease (5 of 14) consistent with breast origin or presented with axillary or regional adenopathy and were found to have metastatic disease on further work-up (9 of 14). Because the striking differences in management of the stage II and stage IV patient groups, we will discuss our findings separately.
Stage II Patients
Breast MRI identified a suspicious lesion in 76% (42 of 55) of stage II patients. Of these 42 suspicious MRI lesions, 26 proved to be the primary breast carcinoma, 12 were false positives, and 4 patients were lost to follow-up after MRI was completed, although biopsy was recommended. The MRI findings and pathologic results for all stage II patients are listed in Table 2.
Of the 26 patients in whom the MRI finding proved to be carcinoma, 19% (5 of 26) of lesions were localized by ultrasonography, 12% (3 of 26) of lesions were localized by MRI, and 57% (15 of 26) of patients did not undergo biopsy before mastectomy, but carcinoma was identified in the same quadrant of the breast in which the MRI abnormality was found. In two patients, when physicians reexamined the patient with knowledge of the MRI findings, they palpated an area of abnormality despite an initial negative physical examination. One patient underwent an excision after the MRI identified a lesion but when MRI localization capability was not yet available.
The dominant tumor type identified by operation or excisional biopsy was ductal carcinoma (76%; 20 of 26). There were two cases of lobular carcinoma, two cases of invasive mammary carcinoma of mixed type, and two cases of ductal carcinoma-in-situ (DCIS). We suspect that microinvasive carcinoma evaded pathologic detection in the two cases of DCIS. Pathologic size data were available for 20 cases (Fig. 1). In six cases, the size was indeterminate. The median size was 1.0 cm (range, .1–2.5 cm).
View larger version (11K):
[in this window]
[in a new window]
|
FIG. 1. Tumor size of invasive cancers identified by magnetic resonance imaging in stage II patients (n = 20).
|
|
Figure 2 demonstrates the surgical treatments for stage II patients in whom the MRI successfully identified the primary lesion. Eleven patients were not breast-conservation candidates because the MRI identified multicentric disease (73%; 8 of 11) or showed extensive disease (27%; 3 of 11). Figures 3 and 4 demonstrate the surgical treatments when the MRI examination was negative (no abnormalities identified; n = 13) and when the MRI examination had a false-positive result (an abnormality on MRI was identified, but biopsy of this lesion did not reveal carcinoma; n = 12). Of the 25 women in whom the MRI did not identify the primary tumor, 12 underwent a mastectomy. The primary carcinoma was found in four of the breast specimens. Two patients with negative MRI examinations had carcinoma identified in their mastectomy specimen. These false-negative MRI examinations included a 3.0-cm invasive ductal carcinoma with DCIS that was missed because the lesion did not demonstrate enhancement (shown in Fig. 5) and a .3-cm invasive lobular carcinoma. Thirteen patients were treated with axillary dissections followed by primary breast irradiation. One patient developed distant disease, nine were without evidence of disease at a median of 4.5 years of follow-up (range, 2–8 years), and three were lost to follow-up. With a median follow-up of 4 years for all stage II patients, 72% (40 of 55) were without evidence of disease, 22% (12 of 55) had developed distant disease, 1.8% (1 of 55) had died of disease, and 3.6% (2 of 55) were lost to follow-up.
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 2. Surgical treatment undertaken in stage II patients (pts) in whom magnetic resonance imaging (MRI) identified a primary breast carcinoma.
|
|
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 3. Treatment undertaken in stage II patients with negative magnetic resonance imaging (MRI) examinations (n = 13). InvD, invasive ductal carcinoma; DCIS, ductal carcinoma-in-situ. InvL, invasive lobular carcinoma; LCIS, lobular carcinoma-in-situ. ALND, axillary lymph node dissection; f/u, follow-up; RT, radiotherapy; NED, no evidence of disease.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 4. Treatment undertaken in stage II patients with false-positive magnetic resonance imaging (MRI) examinations (biopsy of breast lesion did not show carcinoma; n = 12). InvL, invasive lobular carcinoma. InvD, invasive ductal carcinoma; DCIS, ductal carcinoma-in-situ. ALND, axillary lymph node dissection; f/u, follow-up; RT, radiotherapy; NED, no evidence of disease.
|
|
View larger version (177K):
[in this window]
[in a new window]
|
FIG. 5. False-negative breast magnetic resonance image (MRI) that failed to identify a 3.0-cm invasive ductal cancer. Although rare, one limitation of breast MRI is that the lack of vascular enhancement of a neoplasm may lead to an inability to differentiate the tumor from surrounding breast parenchyma. Photograph courtesy of E. A. Morris, Department of Radiology, Memorial Sloan-Kettering Cancer Center.
|
|
Stage IV Patients
In stage IV disease, MRI identified a suspicious breast lesion in 86% (12 of 14) of cases. The MRI-detected lesion was histologically proven to be breast carcinoma in 5 of 9 patients who presented with regional adenopathy and 2 of 5 who presented with distant disease (overall 50%; 7 of 14). There were three false-positive MRI examinations (biopsy of the suspicious MRI lesion did not reveal carcinoma). When the suspicious areas of enhancement on MRI underwent biopsy in two patients, one biopsy revealed benign ductal cells, and the other patient had a fibroadenoma. The last patient had a suspicious fine-needle aspirate of the area of MRI enhancement, but an ultrasound core biopsy of this area revealed stromal fibrosis, apocrine metaplasia, and rare calcifications. Two patients did not undergo biopsy of the MRI abnormality; both had biopsies of their distant metastasis (liver and bone) that were interpreted as "consistent with breast origin," and the MRI finding was viewed as confirmatory.
Mammographic Density
Table 3 lists MRI findings in relation to the mammographic breast density of the entire study population (both stage II and stage IV patients). Most of our patient population was in the mild- to moderate-density Breast Imaging Reporting and Data System categories. There were only two patients with fatty breasts. Breast MRI showed suspicious lesions in all four groups. Whether the breast parenchyma was mild, moderate, or extremely dense, the MRI was equally successful in identifying the primary lesion. Figures 6 and 7 are examples of mammograms of mild and moderate breast density where the occult lesion is easily seen on breast MRI.
View larger version (66K):
[in this window]
[in a new window]
|
FIG. 6. Mammographically occult breast carcinoma visualized by magnetic resonance imaging in a patient with moderately dense breast tissue. Photograph courtesy of E. A. Morris, Department of Radiology, Memorial Sloan-Kettering Cancer Center.
|
|
View larger version (58K):
[in this window]
[in a new window]
|
FIG. 7. Mammographically occult breast carcinoma visualized by magnetic resonance imaging in a patient with mildly dense breast tissue. Photograph courtesy of E. A. Morris, Department of Radiology, Memorial Sloan-Kettering Cancer Center.
|
|
|
DISCUSSION
|
|---|
Although encountered infrequently, occult primary breast cancer remains a formidable diagnostic and therapeutic challenge for both surgeons and oncologists. Similarly challenging is the integration of new technologies into our clinical practice. Whereas breast MRI has shown great promise in select patient groups, those who advocate its widespread use in the general screening population or in patients with demonstrable cancer have been roundly criticized because of a high rate of false positives and a lack of randomized trials. However, patients with occult primary breast cancer in whom the primary tumor cannot be identified by the standard techniques, such as mammography and ultrasonography, seem uniquely suited to this technology. Now that MRI technology has gone through a maturation phase, radiologists have become more experienced in the interpretation of MRI images. In addition, improved image resolution and biopsy capability have vastly improved our ability to act on the imaging findings.
With this in mind, our study supports the routine use of breast MRI in patients with occult primary breast cancer. MRI correctly identified the primary lesion in more than half of our study population. Not only did MRI have the potential to identify patients who may be candidates for breast conservation, but it also correctly identified patients who required mastectomy because of either multicentric or extensive disease. Also, in very select cases, a subtle abnormality on physical examination may be detected once the clinician is directed by the MRI findings.
The other interesting finding, not discussed well in the literature, is the use of breast MRI in patients with stage IV disease. The ability to identify the site of the primary tumor not only can provide an easily accessible area for biopsy, but also can provide additional significant prognostic and therapeutic information. This information may allow for a targeted chemotherapy regimen (i.e., breast vs. lung primary tumor) or act as a marker for response to chemotherapy. The addition of the breast MRI findings for the oncologist seemed to affect the management of stage IV patients; however, it is hard to obtain objective data regarding this inference. Often, the patient underwent a multitude of studies, including computed tomographic scanning and bone scanning, as well as breast MRI, within a few days. However, if the other studies were negative, the diagnosis of metastatic breast cancer was further supported by a positive MRI finding in the breast.
Breast MRI findings can alter surgical or chemotherapeutic treatment. One advantage of breast MRI is that a small percentage of patients may be spared a mastectomy. For those with findings on MRI, MRI identified lesions amenable to breast conservation. Even for those without MRI findings, 10 patients in our series were treated with primary breast irradiation and did not develop breast recurrence with a median follow-up of 4.5 years.
The technique is not without its limitations and still has a percentage of false-positive results. The magnitude of breast density does not seem to limit the efficacy of breast MRI, and, in fact, MRI may reveal occult lesions even in breast tissue that appears mildly dense mammographically. Our experience supports the assertion that occult primary breast cancer is a primary indication for the use of breast MRI. As we continue to gather data about breast MRI—both its advantages and its limitations—we will find further opportunities to use this technology in appropriate clinical settings.
|
FOOTNOTES
|
|---|
Claire L. Buchanan, MD, is now at Swedish Cancer Institute, 500 17th Avenue, Seattle, WA 98108
Received for publication March 4, 2005.
Accepted for publication July 18, 2005.
|
REFERENCES
|
|---|
- Kuhl CK, Schmutzler RK, Leutner CC, et al. Breast MR imaging screening in 192 women proved or suspected to be carriers of a breast cancer susceptibility gene: preliminary results. Radiology 2000;215:267–79.[Abstract/Free Full Text]
- Warner E, Plewes, DB, Shumak RS, et al. Comparison of breast magnetic resonance imaging, mammography, and ultrasound for surveillance of women at high risk for hereditary breast cancer. J Clin Oncol 2001;19:3524–31.[Abstract/Free Full Text]
- Stoutjesdijk MJ, Boetes C, Jager GJ, et al. Magnetic resonance imaging and mammography in women with a hereditary risk of breast cancer. J Natl Cancer Inst 2001;93:1095–102.[Abstract/Free Full Text]
- Tilanus-Linthorst MM, Bartels CC, Obdjein AI, Oudkerk M. Earlier detection of breast cancer by surveillance of women at familial risk. Eur J Cancer 2000;36:514–9.[CrossRef][Medline]
- Morris EA, Liberman L, Ballon DJ, et al. MRI of occult breast carcinoma in a high-risk population. AJR Am J Roentgenol 2003;181:619–26.[Abstract/Free Full Text]
- Olson JA, Jr, Morris EA, Van Zee KJ, Linehan DC, Borgen PI. Magnetic resonance imaging facilitates breast conservation for occult breast cancer. Ann Surg Oncol 2000;7:411–5.[Abstract]
- Morris EA, Schwartz LH, Dershaw DD, et al. MRI imaging of the breast in patients with occult primary breast carcinoma specimens. Radiology 1997;205:437–40.[Abstract/Free Full Text]
- Morris EA. Screening for breast cancer with MRI. Semin Ultrasound CT MR 2003;24:45–54.[CrossRef][Medline]
- Kriege M, Brekelmans CT, Boetes C, et al. Efficacy of MRI and mammography for breast-cancer screening in women with a familial or genetic predisposition. N Engl J Med 2004;351:427–37.[Abstract/Free Full Text]
- Baron PL, Moore MP, Kinne DW, et al. Occult breast cancer presenting with axillary metastases. Updated management. Arch Surg 1990;125:210–4.[Abstract]
- Vlastos G, Jean ME, Mirza AN, et al. Feasibility of breast preservation in the treatment of occult primary carcinoma presenting with axillary metastases. Ann Surg Oncol 2001;8:425–31.[Abstract/Free Full Text]
- Morris EA, Liberman L, Dershaw DD, et al. Preoperative MR imaging-guided needle localization of breast lesions. AJR Am J Roentgenol 2002;178:1211–20.[Abstract/Free Full Text]
- American College of Radiology. Illustrated Breast Imaging Reporting and Data System (BI-RADS). 3rd ed. Reston, VA: American College of Radiology, 1998.
This article has been cited by other articles:
|
|
|
|
 
J. B. Mendel, M. Long, and P. J. Slanetz
CT-Guided Core Needle Biopsy of Breast Lesions Visible Only on MRI
Am. J. Roentgenol.,
July 1, 2007;
189(1):
152 - 154.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. Moy, F. Ponzo, M. E. Noz, G. Q. Maguire Jr., A. D. Murphy-Walcott, A. E. Deans, M. T. Kitazono, L. Travascio, and E. L. Kramer
Improving Specificity of Breast MRI Using Prone PET and Fused MRI and PET 3D Volume Datasets
J. Nucl. Med.,
April 1, 2007;
48(4):
528 - 537.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
