|
|
||||||||
Original Research |
1 Department of Radiology, University of Cincinnati, 234 Goodman St., M.L. 772,
Cincinnati, OH 45267.
2 Department of Pathology, University of Cincinnati, Cincinnati, OH.
3 Department of Surgery, University of Cincinnati, Cincinnati, OH.
Received April 26, 2005;
accepted after revision July 12, 2005.
Address correspondence to M. C. Mahoney.
Abstract
|
|
|---|
MATERIALS AND METHODS. Retrospective review of 1,819 lesions sampled with 11-gauge SVAB yielded 27 patients with lobular neoplasia as the most severe pathologic entity diagnosed. Patients with lobular neoplasia associated with atypical ductal hyperplasia (ADH), ductal carcinoma in situ (DCIS), or infiltrating carcinoma were excluded. Twenty patients underwent surgical excisional biopsy, and seven patients were followed mammographically for a mean of 52 months (range, 14-67 months). Mammographic lesion type, number of specimens obtained per lesion, and specific histologic features related to lobular carcinoma in situ (LCIS) were assessed. Results were compared with histologic findings at surgery or mammographic follow-up.
RESULTS. Nineteen lesions presented mammographically as microcalcifications, four as masses, three as masses with associated microcalcifications, and one as architectural distortion. A mean of 13 specimens were obtained per lesion. Carcinoma was found at surgical excision in 19% of the lesions (5/27). Lesions were upgraded to DCIS (n = 2), invasive lobular carcinoma (n = 2), and mixed invasive ductal and lobular carcinoma (n = 1). In addition to the diagnosis of lobular neoplasia at SVAB, one patient presented with synchronous infiltrating ductal carcinoma in the contralateral breast, and two patients developed metachronous infiltrating ductal carcinoma in a different quadrant of the ipsilateral breast. Twelve of the 27 lesions included LCIS. These lesions were evaluated pathologically to distinguish the classic (10/12) from the pleomorphic (2/12) form of this entity. Ten of the 12 LCIS cases underwent surgical excisional biopsy with four of the five upgrades occurring in these patients. Only one of these patients was shown to have the pleomorphic type of LCIS. Lesions in seven patients who underwent mammographic follow-up remained stable.
CONCLUSION. The known association of lobular neoplasia with high-risk and malignant lesions at surgical biopsy requires careful consideration when lobular neoplasia is diagnosed as the most severe histologic entity at SVAB. The diagnosis of lobular neoplasia at 11-gauge SVAB is not reliable in view of the 19% upgrade rate at the time of surgical excisional biopsy in our study. No predictive mammographic features allowed distinction between the patients with lesions that were upgraded at the time of surgery from those whose lesions were not upgraded.
Keywords: biopsy needle breast cancer lobular neoplasia stereotactic vacuum-assisted biopsy
|
|
|---|
Although most clinicians agree that close clinical and mammographic follow-up is appropriate in cases diagnosed with surgical biopsy, management of these lesions diagnosed at percutaneous biopsy remains controversial. Some investigators advocate surgical excision to exclude sampling error and a more serious pathologic diagnosis. Other investigators point to the fact that because the risk of developing breast cancer involves all breast tissue, surgical excision of one region is not indicated and mammographic follow-up is more appropriate [2-8, 13-16]. Certain parameters for surgical excision have been suggested, which include lesions with features that overlap with those of DCIS, an associated high-risk lesion, mammographic-pathologic discordance, or residual calcifications [3, 7]. Because this entity is infrequently encountered, most published reports have included small numbers of patients, and the lesions were biopsied with a variety of biopsy devices and needle gauges. Our study was designed to determine whether the diagnosis of lobular neoplasia is reliable when identified exclusively with an 11-gauge vacuum-assisted biopsy device. Our hypothesis was that the larger specimens obtained by the 11-gauge vacuum-assisted biopsy device compared with the 14-gauge spring-loaded device and 14-gauge vacuum-assisted biopsy device could provide more accurate histologic sampling and allow mammographic surveillance rather than surgical excision.
|
|
|---|
A retrospective review of our institution's biopsy database was performed by a dedicated breast radiologist. The information obtained included the stereotactic core needle biopsy histologic findings, initial mammographic features of the lesion, lesion size, BI-RADS assessment category, number of cores obtained at biopsy, patient age, presence of residual lesion mammographically after SVAB, and outcome based on surgical histology or mammographic follow-up (Table 1).
|
|
|
|---|
|
|
The seven patients who underwent mammographic surveillance were followed for a mean of 52 months (median, 56 months; range, 14-67 months). No mammographic changes had occurred at the site of SVAB.
In 17 cases, there was no mammographic evidence of residual lesion after SVAB, including three of the five cancers identified at subsequent surgical excision.
The histologic features of the cases of LCIS were retrospectively evaluated and characterized as the classic or pleomorphic type. Ten (83%) of 12 LCIS cases were of the classic form (Figs. 2A and 2B), and two (17%) of 12 were the pleomorphic form (Fig. 3). Ten of these 12 patients underwent surgical excisional biopsy. Four of the five upgrades from surgery occurred in these patients, including one patient with the pleomorphic form of LCIS (Tables 3 and 4).
|
|
|
|
|
|
|
|---|
Lobular neoplasia is frequently multicentric and bilateral. It has been regarded as a highrisk tumor marker, diagnosed primarily in premenopausal women [1]. The estimated risk of developing invasive carcinoma in either breast is approximately five times the rate in the general population. The risk of developing cancer among these patients remains elevated for at least 20 years after the initial diagnosis of lobular neoplasia. The majority of the cancers that develop are invasive ductal carcinomas [9].
Lobular neoplasia is a relatively uncommon finding, with reported rates ranging from 0.8% to 3.8% at breast surgery [1]. At percutaneous core needle biopsy, the reported rates of lobular lesions range from 0.02% to 3.3% [3-8]. In our series, these lesions were found in 1.5% (27/1,819) of all stereotactic core needle biopsies. Liberman et al. [7] have suggested that the incidence of lobular neoplasia detected at percutaneous biopsy will increase with the use of larger needles and vacuum assistance, providing tissue volumes that approach those of surgical biopsy.
In contrast to ductal lesions, lobular neoplasia generally has no clinical or mammographic manifestations [4, 9]. It is usually encountered incidentally at biopsy of a targeted coexistent entity [18]. Recently, however, mammographically evident calcifications have been described in association with lobular lesions [7, 13, 18]. Pathologically, two forms of calcifications are recognized with LCIS [13]. A pleomorphic, necrotic form of LCIS has been described in which the calcifications are associated with necrotic debris and closely resemble the comedocarcinoma calcifications of high-grade DCIS. The classic, nonnecrotic form of LCIS is associated with calcifications that are similar in appearance to those found in benign proliferative change [13]. It has been suggested that the pleomorphic form of LCIS may be associated with a greater likelihood of developing invasive carcinoma than the classic form [13].
The clinical significance of these two forms of LCIS is particularly relevant for patients undergoing stereotactic core needle biopsy. In patients diagnosed with the pleomorphic form of LCIS at stereotactic core needle biopsy, further surgical excision may be warranted to exclude sampling error and underestimation of invasive carcinoma, analogous to the upgrade issues associated with ADH and DCIS [13, 15]. Furthermore, although LCIS and DCIS are distinct entities, their pathologic features can overlap, and histologic differentiation can be difficult. Use of E-cadherin immunohistochemical staining has helped in making the distinction between ductal and lobular neoplasms. Again, this is clinically significant because, as Georgian-Smith and Lawton [13] pointed out, "cases that were previously considered to be DCIS because of necrosis are now being determined to be LCIS (pleomorphic type)."
The experience with LCIS as a high-risk marker rather than a precursor to carcinoma may be based on the exclusion of these more aggressive forms of LCIS, which may have been classified and treated in the past as DCIS. This possibility, coupled with the known association of lobular neoplasia with high-risk and malignant lesions at surgical biopsy [1], requires careful consideration of lobular neoplasia diagnosed as the most severe histologic entity at stereotactic core needle biopsy. In addition, lobular neoplasia may be a part of a spectrum of changes that progress to malignancy, paralleling the changes of ADH that progresses to DCIS, but occurring more slowly [19].
The management of these cases is not well established. Recommendations include excisional biopsy, mammographic follow-up, chemoprevention, or a combination of these approaches [18]. The advantages of percutaneous core needle biopsy compared with surgical excisional biopsy have been well established. These procedures are increasingly being used as the primary means of evaluating suspicious areas of breast tissue. However, because of sampling error inherent in any form of needle biopsy and interpretive errors, which may arise from smaller tissue specimens than those obtained at surgery, the issue of underestimation of malignancy must be addressed. This has been extensively investigated with regard to ADH diagnosed at percutaneous core needle biopsy. Multiple studies have shown that carcinoma can be underdiagnosed as ADH on the basis of findings at core needle biopsy [8, 20]. The current recommendation for a diagnosis of ADH at core needle biopsy is to undergo surgical excisional biopsy to exclude the possibility of coexistent but unsampled DCIS or invasive carcinoma [18]. Similarly, several studies have raised the question as to whether a diagnosis of lobular neoplasia at stereotactic core needle biopsy is a reliable and accurate diagnosis [3-8, 18, 21]. If, as in the case of ADH, the potential exists for a sampling or interpretive error to misdiagnose a more aggressive lesion than lobular neoplasia alone, surgical excision may be necessary. Liberman [17] investigated the percutaneous diagnosis of LCIS and recommended excision "when the percutaneous biopsy histologic features overlapped with those of DCIS, when a high-risk lesion was present, or when there was imaging-histologic discordance." However, a greater diagnostic dilemma exists when lobular neoplasia is diagnosed at percutaneous biopsy and there is no other finding that warrants surgical excision. The question asked in our study is whether the diagnosis of lobular neoplasia is reliable enough in these cases to avoid surgical excision and permit mammographic surveillance as management for these patients. Our findings suggest that this management strategy may not be appropriate.
Previous reports have included both 14- and 11-gauge needles and both spring-loaded and vacuum-assisted devices. We questioned whether the larger tissue specimens provided by the 11-gauge vacuum-assisted biopsy device would provide greater histologic accuracy and allow confident management of lobular neoplasia based solely on the percutaneous biopsy results. However, our study, performed exclusively with an 11-gauge vacuum-assisted biopsy device, still showed upgrades from lobular neoplasia to carcinoma. In fact, three of the five cases in which lobular neoplasia was upgraded to carcinoma at the time of surgical excision occurred in patients with no mammographic evidence of residual lesion after SVAB. So, again, similar to the situation with ADH, although the 11-gauge vacuum-assisted biopsy device may decrease the frequency of upgrades, it does not eliminate this possibility.
In the study by Berg et al. [3], 25 foci of lobular neoplasia (15 ALH, 10 LCIS) were identified in or adjacent to a targeted benign finding. That study viewed patients with residual microcalcifications after percutaneous core needle biopsy as having a higher risk of cancer. Of 11 lesions with residual calcifications, one (9%) was DCIS at subsequent surgery. Excisional biopsy was suggested for cases of lobular neoplasia with the presence of residual microcalcifications after percutaneous core needle biopsy [3]. In a multiinstitutional study by Lechner et al. [5], 34% (20/58) of LCIS cases and 21% (18/84) of ALH cases were associated with malignancy at surgical excision. Irfan and Brem [8] evaluated seven cases of ALH diagnosed at 8-gauge SVAB. Subsequent surgical excision revealed cancer in one case (14%). Shin and Rosen [18] evaluated 14 cases of LCIS and six cases of ALH found on core biopsy. Subsequent excisional biopsy yielded carcinoma in 21% (3/14) of LCIS cases and 17% (1/6) of ALH cases [18]. Similarly, Foster et al. [22] reported the combined results of two institutions and found that 17% of patients with LCIS or ALH at core needle biopsy (35 cases) were upgraded to invasive cancer or DCIS. Cohen [19] combined the data from several of these reported studies and found that 30 (19%) of 159 published reports of ALH, LCIS, or lobular neoplasia were upgraded at the time of excisional biopsy to DCIS or invasive cancer. Dmytrasz et al. [23] reported seven cases of ALH at stereotactic biopsy with three cases of cancer found at surgery. Bauer et al. [24] reported 13 cases of lobular neoplasia without associated cancer or high-risk pathology at core needle biopsy, with one upgrade at excisional biopsy.
These studies reported results that are similar to ours. Although the number of cases in each of the series is small, together, there is a growing body of evidence to support surgical excision in cases of lobular neoplasia identified as the most significant pathologic entity at core needle biopsy.
In our report of 27 patients with lobular neoplasia as the most significant histologic finding at SVAB, there were five cancers (19%) at surgical excision. Because lobular neoplasia is not a commonly encountered entity, as with previous reports, the small number of cases limits our study. Nonetheless, our report represents the largest single institution series and the largest series of cases performed exclusively with an 11-gauge vacuum-assisted biopsy device.
A second limitation of our study is that not all cases had surgical correlation. However, the majority of our cases underwent surgical excision (20/27, 74%). Those with mammographic surveillance (7/27, 26%) had a mean follow-up of greater than 4 years.
All of our cases underwent initial percutaneous biopsy with an 11-gauge vacuum-assisted biopsy device. Despite the greater efficiency of vacuum assistance in obtaining microcalcifications compared with the spring-loaded device and the larger tissue specimens obtained with the 11-gauge needle, we still encountered a 19% incidence of carcinoma at the time of surgery.
We have attempted to address a number of the issues discussed by Dershaw [25] in his recent editorial. Our data add a significant number of cases to the published literature. We included only those cases of lobular neoplasia without coexistent cancer or high-risk disease that would require surgical excision to be performed. In all of our cases, biopsy was performed with an 11-gauge vacuum-assisted biopsy device. The mammographic features of the lobular lesions were evaluated and the histopathologic features of all cases of LCIS.
In conclusion, we recommend that excisional surgical biopsy be performed in cases of lobular neoplasia diagnosed at stereotactic core needle biopsy to exclude a coexistent intraductal or invasive carcinoma that may be present in 19% of these patients.
Acknowledgments
We thank Myron M. Moskowitz for his thoughtful review of the
manuscript.
|
|
|---|
This article has been cited by other articles:
![]() |
G. C. Zografos, F. Zagouri, and T. N. Sergentanis Lobular Neoplasia in Vacuum-Assisted Breast Biopsy: Extended Protocols and Underestimation Am. J. Roentgenol., April 1, 2008; 190(4): W272 - W272. [Full Text] [PDF] |
||||
![]() |
G M Tse, P-H Tan, A L M Pang, A P Y Tang, and H S Cheung Calcification in breast lesions: pathologists' perspective J. Clin. Pathol., February 1, 2008; 61(2): 145 - 151. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S Reis-Filho and S. E Pinder Non-operative breast pathology: lobular neoplasia J. Clin. Pathol., December 1, 2007; 60(12): 1321 - 1327. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |