| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Memorial Sloan Kettering Cancer Center, Weill Medical
College, Office #862, 160 E. 53rd St., New York, NY 10022.
2
Pro Health Radiology, 2800 Marcus Ave., Lake Success, NY 11042.
3
Department of Radiology, Memorial Sloan Kettering Cancer Center, 1275 York
Ave., New York, NY 10021.
4
Breast Oncology Service, Memorial Sloan Kettering Cancer Center, 205 E. 64th
St., New York, NY 10021.
Received October 11, 1999;
accepted after revision February 2, 2000.
Address correspondence to C. B. Winston
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We identified 57 women (age range, 30-79 years; mean age, 57 years) with metastatic lobular carcinoma of the breast who underwent CT of the chest, abdomen, or pelvis between 1995 and 1998. Then two experienced oncology radiologists retrospectively reviewed 78 CT examinations of those patients to identify sites of metastatic disease and to identify complications caused by metastases.
RESULTS. Metastases were identified in bone in 46 patients (81%), lymph nodes in 27 patients (47%), lung in 19 patients (33%), liver in 18 patients (32%), peritoneum in 17 patients (30%), colon in 15 patients (26%), pleura in 13 patients (23%), adnexa in 12 patients (21%), stomach in nine patients (16%), retroperitoneum in nine patients (16%), and small bowel in six patients (11%). Eighteen patients (32%) had gastrointestinal tract involvement that manifested as bowel wall thickening. Hydronephrosis was present in six patients (11%).
CONCLUSION. Although lobular carcinoma metastasized to common metastatic sites of infiltrating ductal carcinoma, lobular carcinoma frequently metastasized to unusual sites, including the gastrointestinal tract, peritoneum, and adnexa. Gastrointestinal tract involvement was as frequent as liver involvement, appearing as bowel wall thickening on CT. Hydronephrosis was a complication of metastatic lobular carcinoma.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The radiologist plays an integral role in examining patients with lobular carcinoma, not only in the detection of the primary lesion but also in the identification of metastatic disease and evaluation of treatment response. With the advent of multiple chemotherapeutic and hormonal agents for the treatment of patients with metastatic breast carcinoma, recognition of metastatic disease and its potential complications has become increasingly important. The CT appearances of metastatic breast carcinoma have been described [4, 5]. However, most studies do not distinguish between patients with lobular carcinoma and those with infiltrating ductal carcinoma and likely reflect patients with infiltrating ductal carcinoma. Because lobular carcinoma has a distinct histology and certain mammographic features that differ from those of infiltrating ductal carcinoma, it is possible that lobular carcinoma also has a different pattern of metastatic spread. Although there have been a few imaging case reports about a small number of patients with metastatic lobular carcinoma [6,7,8], to our knowledge no large series exists describing the relative prevalence of different metastatic sites of lobular carcinoma as seen on CT. Because knowledge of the pattern of disease spread is essential for accurate image interpretation, we undertook this study to determine the distribution of metastatic lobular carcinoma in the chest, abdomen, and pelvis as seen on CT.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One radiologist retrospectively reviewed the reports of all CT examinations obtained for each of the remaining 133 patients. Findings were consistent with or highly suggestive of metastatic disease in 59 patients. Metastases were confirmed at biopsy in 44 of these patients (75%). Metastases were strongly suspected on clinical grounds on the basis of physical examination, elevated tumor markers, or correlative imaging studies (MR images or bone scans) in an additional 13 patients (22%). These 57 patients (age range, 30-79 years; mean age, 57 years) formed our study population. Two other patients (3%) were considered not to have metastatic disease with an alternative diagnosis provided to explain the abnormalities seen on CT (one patient had systemic mastocytosis that was clinically diagnosed, producing diffuse osteosclerosis, and one patient had biopsy-proven Paget's disease of the hip—both patients' conditions were misdiagnosed by the radiologist as metastatic breast carcinoma on CT).
Results of surgical pathologic examination of the primary breast malignancy were available for 53 patients (mastectomy [n = 35], excisional biopsy [n = 14], lumpectomy [n = 4]). Lobular carcinoma was diagnosed in two additional patients with core biopsy and in one patient with fine-needle aspiration. The diagnosis of lobular carcinoma was made in one patient with an axillary lymph node biopsy. The different subtypes of lobular carcinoma previously described [10] were not addressed in this study.
Image Analysis
The CT examinations for each patient were retrospectively reviewed by two
of three experienced oncology radiologists to identify sites of distant
metastatic disease. Decisions were reached by consensus. The most recent CT
examination for each patient was reviewed. Because many patients underwent
multiple imaging studies during their course of treatment, retrospective
review of reports of all CT examinations for each patient was performed to
determine if additional sites of metastatic disease had been identified. If
additional sites were identified on prior examinations, those CT examinations
were retrospectively reviewed as well. A total of 78 CT scans (54 of the
chest, abdomen, and pelvis; 10 of the chest and abdomen; eight of the abdomen
and pelvis; four of the chest; one of the abdomen; and one of the pelvis) were
reviewed (range of CT scans per patient, 1-4; mean number of scans per
patient, 1.4). Five CT examinations (6%) were performed elsewhere and were not
available for retrospective review. The initial reports of those studies were
used in our data collection. The mean time interval between the initial
diagnosis of breast cancer and the CT studies evaluated in this study was 3.3
years (range, 4 days-19 years).
Sixty examinations (77%) were obtained with IV contrast material and 72 (97%) of the 74 abdominal or pelvic CT examinations were obtained with oral contrast material. Effervescent tablets to maximize gastric distention and rectal contrast were not administered. Twenty-four CT examinations (31%) were performed using a helical technique. All CT examinations were retrospectively reviewed at lung and soft-tissue windows. Seventy-two studies (92%) were retrospectively reviewed at liver windows and 68 (78%) at bone windows. The retrospective review of studies performed at other facilities was limited to the windows provided.
The sites of metastatic disease were recorded for each patient. Imaging findings considered suggestive of metastatic disease included lymph nodes larger than 1 cm in short axis (>6 mm for retrocrural lymph nodes); pulmonary nodules, nodular bronchovascular soft tissue, or thickening of the interlobular septa without distortion of the underlying parenchymal architecture; soft tissue along the pleura or pleural enhancement; pericardial soft tissue or pericardial enhancement; liver lesion with heterogeneous enhancement or an infiltrative appearance with distortion of the hepatic vessels; gastric wall thickening or a focal gastric mass in the presence of adequate gastric distention; large- and small-bowel wall thickening or a focal mass in the presence of adequate bowel distention; adrenal mass with an attenuation value exceeding 20 H on unenhanced CT with no visible fat or that had enlarged since a prior study; focal adnexal mass; nodular soft-tissue infiltration along the peritoneum or a focal peritoneal mass; diffuse or nodular soft-tissue infiltration of the retroperitoneal fat; or lytic or sclerotic osseous lesions. Hydronephrosis and bowel obstruction caused by metastatic disease were recorded.
CT scans were retrospectively reviewed with prior imaging studies (when available) for comparison. Only lesions considered suggestive of distant metastatic disease were recorded. Lesions that were too small to characterize or those that remained stable over sequential examinations were not considered metastatic disease. Small cystic ovarian lesions that were thought to represent physiologic cysts were not included. The radiologists knew that pathologic proof or a strong clinical suspicion of metastatic disease was present for each patient in the study, although they did not know the specific metastatic sites.
Histopathologic Confirmation
Pathology records were reviewed to identify the sites of metastatic disease
that were biopsied. The layers of bowel wall involvement were recorded in
patients with histopathologic confirmation of bowel metastases.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
Gastrointestinal tract metastases were seen as frequently as hepatic metastases (n = 18). All patients with gastrointestinal tract involvement on CT had bowel wall thickening. One patient had bowel wall thickening and a focal colonic mass. No patient had an isolated mass involving the bowel wall.
Of 13 patients without pathologic confirmation of metastatic disease, metastases were seen on CT in the following sites: bone (n = 12), lung (n = 6), pleura (n = 6), liver (n = 5), retroperitoneal fat (n = 2), adrenal glands (n = 2), adnexa (n = 1), and pericardium (n = 1). None of the 13 patients had metastatic disease evident in the gastrointestinal tract or peritoneum on CT.
Hydronephrosis caused by metastatic infiltration of the retroperitoneum was identified in six patients (11%). No patient had bowel obstruction on CT.
The sites of biopsy that confirmed the presence of metastatic disease in 44 patients are listed in Table 2. Histopathologic evaluation confirmed metastatic involvement in the bowel wall in nine sites (stomach [n = 3], colon [n = 3], small bowel [n = 2], and appendix [n = 1]) in six patients. Gastric metastases were confirmed in three patients: two had mucosal involvement diagnosed with endoscopic biopsy and one patient had metastatic infiltration of the submucosa, muscularis, and subserosa seen at autopsy. Complete transmural involvement of the colonic wall was identified in two patients (one at colectomy and one at autopsy). Colectomy was performed for bowel obstruction that developed subsequent to the CT examinations included in this study. Colonic mucosal involvement was confirmed in one additional patient with endoscopic biopsy. Metastatic involvement of the small-bowel submucosa, muscularis, and subserosa was seen in two patients (one at autopsy and one at bowel resection). Small-bowel resection was performed for bowel obstruction (which developed subsequent to the CT examinations included in this study). One patient had transmural involvement of the appendix diagnosed at appendectomy performed for presumed appendicitis; no abnormality in the appendix was seen on CT obtained 1 month before surgery.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our series, bone was the most common site of metastatic lobular carcinoma. Although lobular carcinoma also metastasized to lymph nodes, lung, and liver (common metastatic sites for patients with infiltrating ductal carcinoma), it frequently metastasized to the gastrointestinal tract, peritoneum, and adnexa. Gastrointestinal tract involvement was as frequent as liver involvement, appearing as bowel wall thickening on CT. The predilection of lobular carcinoma to metastasize to the gastrointestinal tract, peritoneum, and adnexa seen in this study is in accordance with findings of large autopsy and clinical studies [12,13,14] (Table 3). Interestingly, Harris et al. [12] reported a higher rate of spread of metastatic lobular carcinoma to the peritoneum or retroperitoneum than to bone at autopsy. Lamovec and Bracko [13] reported an equal rate of spread of metastases to peritoneum and bone. Our observation that lobular carcinoma metastasized most frequently to bone likely reflects, in part, the greater sensitivity of CT in the depiction of osseous metastases than of peritoneal and gastrointestinal tract lesions. In our study, CT was performed with routine scanning protocols and was not tailored for the specific evaluation of the gastrointestinal tract. It is possible that if effervescent tablets had been administered to maximize gastric distention and if rectal contrast material had been administered, then even more bowel or peritoneal lesions may have been revealed.
|
Although we did not directly compare the distribution of metastatic lobular carcinoma with that of metastatic infiltrating ductal carcinoma, the autopsy studies that do directly compare these carcinomas [12, 13] report a significantly higher prevalence of spread of metastatic disease to the gastrointestinal tract, peritoneum and retroperitoneum, and ovaries in patients with metastatic lobular carcinoma compared with patients with metastatic infiltrating ductal carcinoma (Table 3). In a clinical study of 2605 patients with invasive breast carcinoma (lobular carcinoma [n = 359] and infiltrating ductal carcinoma [n = 2246]), Borst and Ingold [14] reported a significantly higher prevalence of spread of metastatic lobular carcinoma to the gastrointestinal tract, peritoneum and retroperitoneum, and gynecologic organs than metastatic ductal carcinoma. Metastases in that study were assessed by histologic, radiologic, and physical examination. Taal et al. [15] retrospectively identified 17 patients with metastatic breast carcinoma to the colon or rectum over a 15-year period. Fifteen of these patients (88%) had lobular carcinoma and only one (6%) had infiltrating ductal carcinoma [15]. In another study, Taal et al. [16] identified 27 patients with metastatic breast carcinoma to the stomach during the same time: 20 (74%) had lobular carcinoma and only four (15%) had infiltrating ductal carcinoma. Patients in both of these studies were examined with single or double contrast-enhanced fluoroscopy and endoscopy. The reason for these differences in the metastatic patterns of lobular carcinoma and infiltrating ductal carcinoma is unknown. It has been suggested that loss of expression of the cell-cell adhesion molecule E-cadherin in infiltrating lobular carcinoma may contribute to these differences [17].
In our series, the most common appearance of metastatic lobular carcinoma to the gastrointestinal tract was tumor infiltration along the bowel wall with mural thickening (Figs. 1,2,3). No patient in our study had an isolated mural mass on CT. Although it is sometimes difficult to distinguish serosal implants and true mural involvement on imaging, histopathologic examination confirmed metastatic infiltration within the bowel wall in six patients. Meyers [18] described three different radiologic appearances of hematogenous metastases to the bowel: an intramural mass (as seen in patients with metastatic melanoma), a mesenteric mass with secondary invasion of the bowel wall (as sometimes seen in patients with lung carcinoma), and linitis plastica (bowel wall thickening and rigidity). Linitis plastica, the most common appearance of metastatic breast cancer to the bowel, is usually described in the stomach.
Initial fluoroscopic descriptions of linitis plastica caused by metastatic breast carcinoma do not report the histologic subtype of the primary tumor in most patients [19, 20]. In a subsequent study, Cormier et al. [21] retrospectively identified 31 patients with linitis plastica caused by metastatic breast cancer. All the patients had metastatic lobular carcinoma, and no patient had metastatic infiltrating ductal carcinoma. Harris et al. [12] reported at autopsy that lobular carcinoma metastasized to the stomach in a "diffuse spreading process" and with "a linitis plastica-like appearance in the most severe cases." The cases of gastric metastases caused by infiltrating ductal carcinoma "adopted a nodular configuration." The mural thickening and rigidity described with metastatic breast carcinoma is caused by the dense infiltrate of tumor cells along the bowel wall [18, 19]. This parallels the appearance of lobular carcinoma in the breast. One distinguishing feature of primary lobular carcinoma is the infiltration of malignant cells in single file, often producing an asymmetric density on mammography with no dominant mass [22].
On imaging, linitis plastica caused by metastatic breast cancer may be indistinguishable from primary scirrhous carcinoma of the stomach [20, 23, 24]. Signet cells have been described in both lobular carcinoma and primary gastric carcinoma. A gastric biopsy that reveals adenocarcinoma with signet cell features caused by metastatic lobular carcinoma may be mistaken for a primary gastric carcinoma. The treatment for these two diseases varies greatly. Whereas gastrectomy may be a treatment option for patients with primary gastric carcinoma, systemic chemotherapy or hormonal therapy would be instituted for patients with metastatic lobular carcinoma. Therefore, it is particularly important to be aware of a history of primary lobular carcinoma, because further evaluation with immunohistochemistry may aid in distinguishing these two entities [25].
Patients with genitourinary [26] or gastrointestinal [27, 28] metastases may develop significant complications that warrant intervention. Grabstald and Kaufman [29] reported a series of 24 patients with hydronephrosis from periureteral breast metastases. Asch et al. [27] reported a series of 12 patients with metastatic breast carcinoma to the gastrointestinal tract with complications requiring surgical treatment, including bowel obstruction, gastrointestinal hemorrhage, and pneumoperitoneum caused by a perforated metastasis. Six patients (11%) in our study had hydronephrosis caused by metastatic infiltration of the retroperitoneum. Both large- and small-bowel resections were performed as treatment of intestinal obstruction caused by metastatic lobular carcinoma (which developed subsequent to the CT examinations). One patient in our study underwent appendectomy for symptoms of appendicitis, and histopathologic examination revealed tumor infiltration in the appendiceal wall.
Our institution is a tertiary referral center that treats a large number of patients with breast cancer who present at all disease stages; therefore, we have the opportunity to examine many patients with metastatic lobular carcinoma. It is possible that our data reflect a selection bias for patients with more advanced disease. However, the distribution of metastases in our investigation is in direct accordance with the largest clinical study published thus far evaluating sites of metastatic lobular carcinoma, which included 359 patients with invasive lobular carcinoma (both with and without metastatic disease) over a 17-year period [14] (Table 3).
The limitations of our study should be addressed. Our study was retrospectively performed and reports the various sites of distant metastatic lobular carcinoma on CT. As with any imaging study, we cannot conclude that our results reflect the exact prevalence of metastases to different sites. However, our findings do reflect the results of two major autopsy studies (which theoretically have the highest accuracy in assessing prevalence) [12, 13]. We did not assess for leptomeningeal carcinomatosis, a finding described with metastatic lobular carcinoma [11], because CT images were obtained with a large field of view for the evaluation of the chest, abdomen, and pelvis. Chest wall recurrences are usually clinically evident, and CT scans are not routinely obtained to detect local tumor recurrence. Therefore, we reviewed CT scans for distant metastases only. Because biphasic contrast-enhanced CT scans were not routinely obtained, we did not characterize the enhancement pattern of hepatic metastases as hypervascular or hypovascular.
The purpose of this study was not to assess the sensitivity or specificity of imaging in the detection of metastatic breast carcinoma, but rather to determine the distribution of metastatic lobular carcinoma. Although we do not have pathologic proof for all sites thought to represent metastatic disease on CT, pathologic confirmation of metastatic breast cancer was available for at least one site per patient in 44 patients (77%). In clinical practice, once metastatic disease is documented in a patient, biopsy of additional lesions is usually not required. Strong clinical suspicion of metastatic disease existed in an additional 13 (23%) patients on the basis of physical examination, elevated tumor markers, and correlative imaging studies. Most metastases in this group of 13 patients were in the bone, lung, pleura, and liver. None of these 13 patients had metastases in the gastrointestinal tract or peritoneum on CT. We purposefully included these 13 patients to avoid a selection bias for patients with atypical metastatic sites that required biopsy. We excluded patients who had a second primary malignancy to ensure that the metastatic lesions revealed on imaging were from lobular carcinoma. The authors recognize, as with all imaging studies, radiologic review is not equivalent to proven metastatic disease.
We did not correlate the specific sites of metastatic disease with the time elapsed since different types of therapy. It is possible that various types of chemotherapy, hormonal therapy, or immunosuppressive therapy (bone marrow transplantation) may alter the pattern of metastatic disease. However, a large-scale study evaluating the precise relationship of therapy to the onset of metastases at various sites with CT scans obtained at regular time intervals would be required to accurately address this question.
In conclusion, although lobular carcinoma metastasized most frequently to bone, lobular carcinoma has a propensity to metastasize to the gastrointestinal tract, peritoneum, and adnexa (relatively atypical sites for patients with infiltrating ductal carcinoma). Gastrointestinal tract involvement was as frequent as hepatic involvement in this series, appearing as bowel wall thickening on CT. The radiologist should be aware of the histologic subtype of breast carcinoma when interpreting body CT for metastatic disease. Knowledge of this pattern of disease spread will not only aid in the detection of metastatic disease and its potential complications, but also aid in minimizing the possibility of mistaking metastatic disease for a second primary malignancy.
Acknowledgments
We thank Hyok-Hee Yoo for his much appreciated assistance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M Brookes, D MacVicar, and J Husband Metastatic carcinoma of the breast: the appearances of metastatic spread to the abdomen and pelvis as demonstrated by CT Br. J. Radiol., April 1, 2007; 80(952): 284 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Harvey Unusual Breast Cancers: Useful Clues to Expanding the Differential Diagnosis Radiology, March 1, 2007; 242(3): 683 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Y. Kim, K. W. Kim, A. Y. Kim, H. K. Ha, J.-S. Kim, S. H. Park, J. K. Kim, M.-J. Kim, S. W. Park, and M.-G. Lee Bloodborne metastatic tumors to the gastrointestinal tract: CT findings with clinicopathologic correlation. Am. J. Roentgenol., June 1, 2006; 186(6): 1618 - 1626. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. C. McLemore, B. A. Pockaj, C. Reynolds, R. J. Gray, J. L. Hernandez, C. S. Grant, and J. H. Donohue Breast Cancer: Presentation and Intervention in Women With Gastrointestinal Metastasis and Carcinomatosis Ann. Surg. Oncol., November 1, 2005; 12(11): 886 - 894. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. El Khoury, P. Cherel, V. Becette, C. De Maulmont, V. Costes, V. Talma, and C. Hagay Unusual Soft-Tissue Metastasis of an Invasive Lobular Carcinoma Mimicking Fasciitis Am. J. Roentgenol., March 1, 2004; 182(3): 745 - 747. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sheth, K. M. Horton, M. R. Garland, and E. K. Fishman Mesenteric Neoplasms: CT Appearances of Primary and Secondary Tumors and Differential Diagnosis RadioGraphics, March 1, 2003; 23(2): 457 - 473. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
