Wo m e n ’s I m a g i n g • O r i g i n a l R e s e a r c h Ha et al. Percutaneous Biopsy of Mucocele-Like Lesions in the Breast
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
Women’s Imaging Original Research
Mucocele-Like Lesions in the Breast Diagnosed With Percutaneous Biopsy: Is Surgical Excision Necessary? Daon Ha1 Vandana Dialani2 Tejas S. Mehta2 Whitney Keefe1 Elaine Iuanow 2,3 Priscilla J. Slanetz 2 Ha D, Dialani V, Mehta TS, Keefe W, Iuanow E, Slanetz PJ
Keywords: biopsy, breast, mucocele-like lesion DOI:10.2214/AJR.13.11988 Received September 28, 2013; accepted after revision May 14, 2014. 1
Tufts University School of Medicine, Boston, MA.
2 Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, Shapiro Clinical Center, 4th Fl, 330 Brookline Ave, Boston, MA 02215. Address correspondence to P. J. Slanetz (
[email protected]). 3
Present address: Salt Lake City, UT.
This article is available for credit. AJR 2015; 204:204–210 0361–803X/15/2041–204 © American Roentgen Ray Society
204
OBJECTIVE. The purpose of this study was to determine the frequencies of atypia and cancer at excisional biopsy of lesions with a diagnosis of mucocele-like lesion (MLL) at percutaneous breast biopsy. MATERIALS AND METHODS. Retrospective review of 9286 lesions subjected to percutaneous imaging-guided biopsy identified MLLs in 35 (0.38%) patients. Medical records, imaging studies, and histologic results were reviewed. RESULTS. Of the 35 patients with core biopsy findings of MLL, 27 underwent stereotactic core needle biopsy (19 with microcalcifications, five with calcifications with an associated mass, and three with only a mass), and eight underwent ultrasound-guided core needle biopsy (four with a solid mass, three with a complex cystic mass). At core biopsy, 12 of 35 (34%) MLLs were associated with atypia (10 cases of atypical ductal hyperplasia, two of flat epithelial atypia), and 23 of 35 (66%) were benign MLL only. All 12 MLLs associated with atypia and 12 of 23 benign MLLs were surgically excised. Eleven patients did not undergo surgery, five of whom were lost to follow-up. One of the 12 (3% of the 35) MLLs associated with atypia was upgraded to DCIS. None were upgraded to invasive cancer. None of the benign MLLs were upgraded to malignancy, and findings at excision of four of the 23 (17%) benign MLLs led to a change in diagnosis to a high-risk lesion (three atypical ductal hyperplasia, one atypical lobular hyperplasia). CONCLUSION. MLL is a rare diagnosis but is encountered in large-volume breast practices. The findings are nonspecific with a range of imaging appearances. No imaging test is reliable for differentiating MLL from other suspicious lesions or lesions with associated atypia. Surgery is clearly warranted for MLL associated with atypia at core needle biopsy because it may be upgraded to malignancy upon excision. However, if the presence of atypia at excision of benign MLL will change clinical management, then benign MLL at core needle biopsy warrants surgical excision in some cases. In patients whose treatment will not change if atypia is found at excision, close surveillance with short-interval follow-up is a reasonable alternative.
M
ucocele-like lesion (MLL) of the breast is a rare lesion characterized by dilated epithelium-lined ducts filled with mucin, often associated with extravasation of mucin into the stroma [1]. Although MLL was originally described as a benign lesion [2], further research has shown that the epithelium lining the ducts may have an array of proliferative changes ranging from atypical ductal hyperplasia (ADH) to ductal carcinoma in situ (DCIS) [1]. It has been suggested that MLL may be a precursor lesion in the spectrum of pathologic lesions that includes mucinous DCIS and invasive mucinous carcinoma [3]. Because of the association between MLL and atypical ductal proliferative lesions and
the difficulty of diagnosis by core needle biopsy (CNB) due to the limited nature of the sample, breast pathologists typically recommend surgical excision whenever MLL is found at CNB [4]. However, because MLL is a rare finding, relatively few data exist on the rate of malignancy associated with MLL, leaving it unclear whether the current recommendation is truly warranted [5]. Likewise, few studies have been conducted on the imaging findings of MLL to help detect and guide its further management. The primary purpose of our study was to collect additional data on the frequency of associated cancer in lesions with a diagnosis of MLL at percutaneous core biopsy to guide management recommendations. A secondary purpose was
AJR:204, January 2015
Percutaneous Biopsy of Mucocele-Like Lesions in the Breast
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
to review the imaging findings of MLL and to identify features that may correlate with the histopathologic findings.
TABLE 1: Frequency of Atypia in Mucocele-Like Lesions at Core Needle Biopsy Diagnosis and Relation to Various Clinical and Imaging Characteristics
Materials and Methods The study was approved by the institutional review board at our hospital, which is an academic center of a major teaching hospital, and informed consent from patients was deemed unnecessary. A retrospective review of the hospital’s online medical records identified all percutaneous biopsies performed in the 8-year period from January 1, 2006, through December 31, 2013. Patient demographic, imaging, and follow-up data were obtained for patients with a percutaneous CNB result of MLL. Information collected included patient age, risk factors for breast cancer (family history or personal history of previous breast cancer or atypia), lesion characteristics, and modality of guided biopsy (stereotactic or ultrasound). Number of specimens obtained and needle gauge were also recorded when available. Clinical and imaging follow-up information and treatment were recorded from the online medical records. Length of imaging follow-up since initial biopsy was noted for all patients, including those who did not undergo surgical excision. Imaging features were reviewed by two breast fellowship–trained radiologists with more than 7 years’ experience. In all cases the initial pathologic analysis was provided by breast pathologists as cases were assigned to them by department protocol. The pathology slides were subsequently also reviewed by a breast pathologist for confirmation of diagnosis. Any cases with change in diagnosis from CNB pathologic analysis to surgical pathologic analysis, including benign to atypia, benign to DCIS, and atypia to DCIS, were considered upgrades.
Characteristic
No. of Cases of Atypia or DCIS/ Total No. of Mucocele-Like Lesions
Age (y) (n = 35) ≤ 52
6/19
> 52
6/16
Family history of breast cancer (n = 29) Yes
1/2
No
11/27
Previous breast atypia or DCIS (n = 15) Yes
3/3
No
8/12
Menopausal status (n = 35) Premenopausal
4/14
Perimenopausal
3/5
Postmenopausal
5/16
Mammographic appearance (n = 34) Microcalcification
10/19
Mass
1/3
Microcalcification and mass
1/5
Asymmetry
0/7
Sonographic appearance (n = 16) Not seen
1/7
Cystic lesion
2/4
Solid lesion
1/5
Size of lesion on mammogram (n = 28) ≤ 5 mm
10/21
> 5 mm
2/7
Note—DCIS = ductal carcinoma in situ.
Biopsy Methods and Tissue Processing All CNBs were performed by board-certified fellowship-trained breast radiologists (1–15 years of experience). The tissue specimens were formalin-fixed and processed according to the 2007 College of American Pathologists/American Society of Clinical Oncology guidelines. Three 5-µmthick sections were examined for each paraffinembedded tissue block according to the standard microtomy protocol for breast CNB at our institution. In 27 cases, biopsy was performed under stereotactic guidance with an 8-, 9-, or 11-gauge vacuum-assisted biopsy device. The median number of specimens obtained at stereotactic biopsy was eight (range, 4–21). In the other eight cases, biopsy was performed under ultrasound guidance with a 14-gauge spring-loaded needle in seven cases and with an 11-gauge vacuum-assisted biopsy device in one case. The median number of specimens obtained was six (range, 5–6).
Imaging and Clinical Follow-Up Among 24 lesions with available surgical reports, surgery was performed a median of 9 weeks (range, 3–15 weeks) after percutaneous biopsy. The median interval for surgery was shorter when MLL was associated with atypia (range, 4–10 weeks) and longer when CNB showed benign MLL (range, 3–15 weeks). Follow-up mammograms after surgery were available for 29 patients (median follow-up duration, 2 years; range, 6 months–6 years). None of the patients had a recurrence. Eleven patients did not undergo surgery. Five of these patients were lost to follow-up. Follow-up data collected for 1–6 years were available for the other six patients and did not show any cancer at the site of biopsy. At our institution, all patients with a diagnosis of ADH, atypical lobular hyperplasia (ALH), or lobular carcinoma in situ undergo evaluation in our breast care center. If the
lesions are deemed high risk, the patients are offered chemoprevention.
Statistical Analysis Statistical analysis was performed with statistical software (Stata 13.1, StataCorp) and Microsoft Excel 2007. The statistical significance of the differences in frequency of change in diagnosis or upgrade at surgical biopsy and the relation to biopsy needle and lesion characteristics was calculated with the Fisher exact test. The Student t test was used to analyze the difference between the ages of patients with atypia or DCIS with MLL and patients with benign MLL. Two-tailed p < 0.05 was considered statistically significant.
Patients and Lesions Of the 9286 lesions biopsied percutaneously under stereotactic or sonographic guidance
AJR:204, January 2015 205
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
Ha et al. during the study period, 35 (0.38%) were diagnosed as MLL. The lesions were found in 35 patients (all women; median age, 52 years; range, 23–84 years). Fourteen women were premenopausal, five were perimenopausal, and 16 were postmenopausal. Three patients had a family history of breast cancer in first-degree relatives, and three patients had a personal history of previous atypia or DCIS. Two had a history of atypia or DCIS in the breast contralateral to the breast that had the MLL. One patient had ADH in the ipsilateral breast diagnosed 10 days before MLL was diagnosed. The first CNB corresponding to one cluster of microcalcifications showed ADH and was 2 cm away from the CNB for calcifications corresponding to MLL. The subsequent surgical excision included both biopsy sites, and the histologic result was DCIS with no residual MLL. After careful review of the histopathologic findings by a breast pathologist, the DCIS was deemed associated with the initial core biopsy site that showed ADH and not with the biopsy site that revealed MLL. This case was therefore considered benign MLL with no upgrade in diagnosis at surgery. This was the patient who presented with bloody nipple discharge, which could have been due to DCIS in the same breast. Analysis of the findings on the 35 MLL patients who had surgical pathologic results revealed no specific clinical risk factors contributing to the frequency of atypia or change in diagnosis. The frequency of atypia with MLL found with CNB was comparable to that of benign MLL and had no statistically significant association with patient age, family history, past breast pathologic findings, menopausal status, appearance at imaging, or size of the lesion (Table 1). The frequency of change in diagnosis at surgical excision had no statistically significant association with biopsy needle size, number of cores obtained, size of the lesion, or whether removal of MLL at percutaneous biopsy was complete (Table 2). Of the 35 patients, 31 had clinically occult lesions, three presented with a palpable mass, and one had bloody nipple discharge in the setting of DCIS in the ipsilateral breast. Findings from diagnostic imaging preceding percutaneous biopsy were available in all cases (19 mammography only, 15 mammography and ultrasound, one ultrasound only). Thirty-one patients underwent workup at our institution. The other four were referred for biopsy at our institution after initial workup, but their images were available for review. Surgical pathologic results were available for only 24 of the 35 patients, although surgical excision was recommended for all patients as a standard of care, given the relative paucity of data.
206
TABLE 2: Frequency of Changes in Diagnosis at Surgical Excision From Percutaneous Biopsy and Relation to Biopsy Needle and Lesion Characteristics No. of Upgradesa/ Total No. of Mucocele-Like Lesionsb
Characteristic Needle size (n = 24) 8- or 9-gauge vacuum
4/19
11-gauge vacuum
0/3
14-gauge spring-loaded
1/2
No. of specimens (n = 24) ≤5
1/4
6–10
3/14
≥ 11
1/6
Size of lesion on mammogram (n = 24) ≤ 5 mm
3/21
> 5 mm
2/3
Mucocele-like lesion seen in surgical pathologic analysis (n = 24) Yes
14/24
No
10/24
Note—p > 0.05 for biopsy needle and all lesion characteristics. aNumber of changes in diagnosis from percutaneous biopsy to surgical pathologic analysis, including benign to atypia, benign to ductal carcinoma in situ, or atypia to ductal carcinoma in situ. bOnly cases with surgical pathologic results.
Results Among 35 patients with a diagnosis of MLL at CNB, 27 underwent stereotactic CNB, and eight underwent ultrasound-guided CNB. Review of the imaging findings revealed that 19 MLLs presented as calcifications on mammograms (Fig. 1), five had calcifications with an associated mass, and three had only a mass. The median size of the abnormality on mammograms was 8 mm (n = 27; range, 3–20 mm). In 15 of the 19 cases, the calcifications were associated with MLLs; in four cases, however, MLLs were associated with other benign breast disease (columnar cell change in three cases and benign breast tissue in one case). Eight patients underwent ultrasound-guided CNB. Three of these patients had a complex cystic mass, one had a solid mass, and three had a mass with calcifications (Fig. 1). The median size of the abnormality at ultrasound was 7 mm (n = 8; range, 2–12 mm). All cases were BI-RADS category 4, and percutaneous biopsy was recommended. At percutaneous core biopsy, 12 of 35 (34%) MLLs were associated with atypia (10 ADH, two flat epithelial atypia [FEA]), and 23 of 35 (66%) were benign MLLs. All 12 MLLs associated with atypia at CNB were removed by surgical excision. One of the 12 (8%) cases of atypia was upgraded to DCIS at surgery.
The excision specimen in this upgraded case contained a minute focus of DCIS adjacent to the biopsy site. The case with a final diagnosis of DCIS is summarized in Figure 2. Ten of the 12 MLLs with associated atypia were found to have additional atypia at excision and no change in diagnosis. One patient with the CNB finding of focal ADH had benign pathologic results at excision. Figure 3 summarizes the pathologic results of CNB and surgical excision in these cases. Of the 23 MLLs found benign at CNB, 12 (52%) were surgically excised, and 11 (48%) were not. Four of 12 (33%) MLLs with benign histologic findings at CNB had changes in diagnosis to high-risk lesions at surgery (three cases of ADH, one case of ALH). Among the 11 (48%) MLLs not surgically excised, five (22% of 23) were lost to follow-up. The other six (26%) patients underwent follow-up mammography for 1–4 years with no signs of interval cancer. Three underwent 6-month follow-up mammography for 2 years, had no interval cancer, and resumed annual screening. As of this writing, three patients were undergoing follow-up at 6-month intervals and will be followed according to departmental protocol for BIRADS category 3 findings. Among 24 excised lesions (12 benign MLLs, 12 MLLs with atypia) for which path-
AJR:204, January 2015
Percutaneous Biopsy of Mucocele-Like Lesions in the Breast TABLE 3: Summary of Upgraded Cases
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
Imaging
Biopsy
Patient No.
Age (y)
Presentation
1
38
Palpable mass
Cluster of heterogeneous 3-mm calcifications, BI-RADS 4
Not performed
2
52
Callback for asymmetry at screening
Macrolobulated mass, BI-RADS 4
Lobulated well- circumscribed hypoechoic 1.2-cm mass
3
51
Screening, asymptomatic
4
50
5
48
Surgical Pathologic Finding
No. of Cores
Pathologic Finding
Stereotactic 9-gauge vacuum-assisted probe
6
MLL with calcifications
ADH
Ultrasound-guided 14-gauge spring loaded
6
MLL with calcifications
ADH
Clustered microcal Hypoechoic irregular cifications associated with 4-mm mass 5-mm mass, BI-RADS 4
Stereotactic 9-gauge vacuum-assisted probe
8
MLL with calcifications
ADH
Screening, asymptomatic
Cluster of pleomorphic 4-mm microcalcifications, BI-RADS 4
Not performed
Stereotactic 9-gauge vacuum-assisted probe
11
MLL with calcifications
ADH
Screening, asymptomatic
Cluster of linear 6-mm microcalcifications, BI-RADS 4
Not performed
Stereotactic 9-gauge vacuum-assisted probe
6
MLL with ADH
DCIS
Mammography
Ultrasound
Method
Note—MLL = mucocele-like lesion, ADH = atypical ductal hyperplasia, DCIS = ductal carcinoma in situ.
ologic findings were available, 14 (40% of the 35 MLLs found at CNB) were associated with atypia, one was associated with DCIS (3%), and nine were benign (26%) at final diagnosis. Of the 24 lesions that were excised, 10 had no evidence of MLL at surgical excision, but at excision, biopsy site changes with other pathologic findings were noted: three cases of columnar cell change, three cases of unremarkable breast tissue, two cases of microcysts,
one case of FEA, and one focus of DCIS. Five patients had an upgrade (one, atypia to DCIS; four, benign MLL to atypia).Thus the upgrade rate of benign MLLs was 17% (4/24). A more detailed description of the upgraded cases is shown in Table 3. Of the 14 patients with atypia (ADH or ALH), seven (50%) were counseled about risk reduction chemoprevention. Only two of the seven (28%) agreed to chemoprevention
A
and started tamoxifen treatment. Six of the 14 (43%) had undergone previous chemoprevention and counseling because of known atypia. One of the 14 (7%) was not counseled about having the surgery performed at our institution and was subsequently lost to follow-up. Discussion MLL is a rare diagnosis among all breast diseases. In our study, MLL was found in
B
Fig. 1—Mucocele-like lesions presenting as calcifications. A, 46-year-old woman undergoing screening mammography. Magnified craniocaudal view shows cluster of indeterminate microcalcifications (circle). Stereotactic core needle biopsy result was benign mucocele-like lesion associated with microcalcifications. Subsequent surgical excision revealed residual mucocele-like lesion. B, 76-year-old woman with palpable lesion. Ultrasound image shows complex cystic lesion (arrows). Subsequent core needle biopsy showed mucocele-like lesion with associated focal atypical ductal hyperplasia and flat epithelial atypia. Subsequent surgical excision also revealed mucocele-like lesion with atypical ductal hyperplasia arising in background of flat epithelial atypia.
AJR:204, January 2015 207
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
Ha et al.
A
B
C
D
Fig. 2—48-year-old woman with diagnosis upgraded from mucocele-like lesion associated with atypia to ductal carcinoma in situ at surgical excision. A, Mediolateral oblique mammogram of left breast shows 6-mm cluster of irregular calcifications (circle). B, Mediolateral magnification mammogram shows calcifications. C, Photomicrograph (H and E, ×200) of core needle biopsy specimen shows mucocele-like lesion, which consists of dilated mucin-filled ducts lined by monomorphic population of atypical cells and associated stromal mucin extravasation. (Courtesy of Singh K, Beth Israel Deaconess Medical Center, Boston, MA) D, Photomicrograph (H and E, ×200) of surgical specimen shows minute focus of ductal carcinoma in situ (low nuclear grade with cribriform pattern) adjacent to biopsy site. No residual mucocele-like lesion was seen in surgical specimen. (Courtesy of Singh K, Beth Israel Deaconess Medical Center, Boston, MA)
0.38% of percutaneous breast biopsies, similar to the previously reported rate of 0.25% [5]. Among 24 patients with surgically excised MLLs, one patient (3%) was found to have DCIS at excision, and 14 patients (40%) had associated high-risk lesions (ADH, ALH, or FEA). These findings are similar to those in previous studies showing a 46–70% incidence of ADH, DCIS, or invasive carcinoma [6–8]. In our study, four of 23 (17%) MLLs benign at CNB had changes at excision to a diagnosis of atypia. In previous studies, the rate of change from a benign MLL diagnosis at percutaneous biopsy to ADH or malig-
208
nancy at excision ranged between 18% and 43% [5, 6, 9]. Most of the studies reported in the radiology literature have small datasets of 10 or fewer patients. In the pathology literature, Jaffer et al. [5] reported that eight of 45 (18%) patients with a benign MLL diagnosis at CNB had changes in diagnosis to ADH or intraductal carcinoma at excision. The rate of diagnosis change in our study was similar to that observed by Jaffer et al. Several studies have shown that the risk of upgrade at surgery is strongly associated with the presence of atypia at CNB and that MLL without atypia diagnosed at CNB
is usually associated with a benign outcome. Sutton et al. [10] found that all 22 MLLs without atypia were benign at excision but that five of 16 (31%) cases of atypia at CNB were upgraded to DCIS at excision. Edelweiss et al. [11] reported upgrade in four of 18 (22%) cases of MLL with atypia at CNB [11] but that no upgrade was noted in 10 cases of MLL without atypia at CNB. Rakha et al. [12] found an upgrade in only two of 54 (4%) cases of MLL with no atypia at CNB. In our study, the rate of upgrade to malignancy of MLL with atypia was 8% (1/12) and of benign MLL was 0% (0/12). Although our
AJR:204, January 2015
Percutaneous Biopsy of Mucocele-Like Lesions in the Breast
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
MLL on CNB (n = 35) Benign MLL (n = 23)
MLL with atypia (n = 12) No surgery (n = 0)
Surgical excision (n = 12) DCIS (n = 1)
Atypia (n = 10)
Benign (n = 1)
Surgical excision (n = 12) DCIS (n = 0)
Atypia (n = 4)
No surgery (n = 11) Benign (n = 8)
Follow-up; no interval cancer (n = 6)
Lost to follow-up (n = 5)
Fig. 3—Chart shows pathologic results after core needle biopsy (CNB) and surgical excision. Thirty-five mucocele-like lesions (MLLs) were diagnosed at CNB, 12 of which were associated with atypia. All 12 MLLs associated with atypia were removed by surgical excision, which revealed one upgrade to ductal carcinoma in situ (DCIS). Twelve of 23 MLLs found benign at CNB were removed by surgical excision; four of the 12 excised MLLs had change in diagnosis to atypia.
finding is not statistically significant because of the sample size of this rare disease (p = 0.5), it is in accordance with these three studies: MLL associated with atypia is more likely to be upgraded to malignancy at excision than is benign MLL. The authors of the previous studies argued that because the rate of upgrade is low in cases of benign MLL, surgical excision may not be necessary when MLL is benign at CNB. In other words, although it is prudent to proceed with surgical excision when MLL with atypia is diagnosed at CNB, for benign MLL, vigilant follow-up may be sufficient. Although we agree with this suggestion, there are additional factors to take into consideration. In our study, we identified a 33% change in diagnosis (four atypical lesions in 12 benign MLL excised and in a total of 23 benign MLL lesions) from benign MLL at CNB to atypia after excision. The finding of atypia can be important in some cases when it changes a patient’s treatment. For example, some patients may choose to take chemopreventative agents, such as tamoxifen or an aromatase inhibitor, to reduce the risk of subsequent malignancy based on personal risk factors, including a biopsy showing atypia. Therefore, surgical excision may be warranted after benign MLL is found at CNB in cases in which the diagnosis of atypia would inform clinical decision making. In contrast, for women who already have a diagnosis of atypia, finding additional atypia would not alter their treatment, and therefore surgery could safely be avoided. These patients can undergo 6-month follow-up imaging for 2–3 years. For patients who defer excision, follow-up imaging at 6, 12, 24, and 36 months is reasonable. Vacuum-assisted needle biopsy has been found to have a lower false-negative rate
than spring-loaded CNB [13]. Use of the vacuum-assisted device has been suggested as the diagnostic technique of choice for lesions unlikely to prove malignant and to reduce the need for surgical excision in some cases with careful follow-up [9]. In our study, most of the biopsies were performed with a vacuum-assisted device, but the frequency of false-negative diagnoses was still considerably high, as evidenced by the frequency of changes in diagnosis at surgery. Even in cases in which MLL was completely removed at biopsy, as evidenced by the absence of MLL at surgical pathologic analysis, there were changes in diagnosis at surgical excision. This finding suggests that even with successful target removal at percutaneous biopsy with direct imaging visualization, the presence of associated atypia or malignancy is still possible. Clearly, use of percutaneous biopsy of MLL may cause one to miss the diagnosis of associated atypia even when the targeted MLL is completely removed. This observation also supports the argument for surgical excision or at least close follow-up when MLL is diagnosed at CNB. Most MLLs are identified as indeterminate calcifications on mammograms [3]. Kim et al. [14] reported that the mammographic appearance of MLL was characterized by the presence of pleomorphic calcifications and that in malignant MLLs, microcalcifications extended over a wider area than did benign MLLs. In our study, calcifications were also the most common mammographic finding. The sonographic appearance of MLL most commonly reported is a cystic mass [15]. In our study, however, only three of eight masses presented as cystic masses at ultrasound imaging. Differences in imaging character-
istics between malignant and benign MLLs could not be determined because of the small number of malignancies in our study. No apparent differences were appreciated between the imaging findings of benign MLL and MLL with atypia. Most of the patients (78% [14/18]) had no symptoms, and the MLLs were detected at screening mammography. There has been a disagreement in the literature about the relation between patient age and the risk of malignancy associated with MLL. One study [16] showed that among patients with MLL, those with a malignant form were older (mean age, 67 vs 59 years). The study, however, included only 10 patients, and the findings were not statistically significant. In another study, Kim et al. [14] had opposite findings: Patients with MLLassociated malignancy tended to be younger (mean age, 37.2 vs 41.4 years), and the findings were statistically significant (p < 0.05). Like Kim et al., we found that the women who had atypia with MLL were younger (mean age, 51 vs 57 years), and the only patient with an upgrade to DCIS was relatively young (48 years). Conclusion MLL is a rare diagnosis but is encountered in large-volume breast practices. There is no definitive test for differentiating MLL from other suspicious lesions or for identifying MLLs with associated atypia. If the presence of atypia at excision of benign MLL will change clinical management, then benign MLL found at CNB warrants surgical excision in some cases, because atypia is relatively commonly discovered at surgical excision (33%). In patients whose clinical care will not change, surveillance with close follow-up imaging is a reasonable approach.
AJR:204, January 2015 209
Downloaded from www.ajronline.org by SUNY Downstate Medical Center on 03/21/15 from IP address 138.5.159.110. Copyright ARRS. For personal use only; all rights reserved
Ha et al. Acknowledgments We thank Kamaljeet Singh for reviewing and providing histologic images and Laura Collins for comments on an earlier draft of this paper. References 1. Schnitt SJ, Collins LC. Biopsy interpretation of the breast. Philadelphia, PA: Lippincott Williams & Wilkins, 2009 2. Rosen PP. Mucocele-like tumors of the breast. Am J Surg Pathol 1986; 10:464–469 3. Leibman AJ, Staeger CN, Charney DA. Mucocelelike lesions of the breast: mammographic findings with pathologic correlation. AJR 2006; 186:1356–1360 4. Jacobs TW, Connolly JL, Schnitt SJ. Nonmalignant lesions in breast core needle biopsies: to excise or not to excise? Am J Surg Pathol 2002; 26:1095–1110 5. Jaffer S, Bleiweiss IJ, Nagi CS. Benign mucocelelike lesions of the breast: revisited. Mod Pathol
2011; 24:683–687 6. Begum SM, Jara-Lazaro AR, Thike AA, et al. Mucin extravasation in breast core biopsies: clinical significance and outcome correlation. Histopathology 2009; 55:609–617 7. Carkaci S, Lane DL, Gilcrease MZ, et al. Do all mucocele-like lesions of the breast require surgery? Clin Imaging 2011; 35:94–101 8. Ohi Y, Umekita Y, Rai Y, et al. Mucocele-like lesions of the breast: a long-term follow-up study. Diagn Pathol 2011; 6:29 9. Carder PJ, Murphy CE, Liston JC. Surgical excision is warranted following a core biopsy diagnosis of mucocoele-like lesion of the breast. Histopathology 2004; 45:148–154 10. Sutton B, Davion S, Feldman M, Siziopikou K, Mendelson E, Sullivan M. Mucocele-like lesions diagnosed on breast core biopsy: assessment of upgrade rate and need for surgical excision. Am J Clin Pathol 2012; 138:783–788 11. Edelweiss M, Corben AD, Liberman L, et al. Fo-
cal extravasated mucin in breast core needle biopsies: is surgical excision always necessary? Breast J 2013; 19:302–309 12. Rakha EA, Shaaban AM, Haider SA, et al. Outcome of pure mucocele-like lesions diagnosed on breast core biopsy. Histopathology 2013; 62:894–898 13. Jackman RJ, Marzoni FA Jr, Rosenberg J. Falsenegative diagnoses at stereotactic vacuum-assisted needle breast biopsy: long-term follow-up of 1280 lesions and review of the literature. AJR 2009; 192:341–351 14. Kim JY, Han BK, Choe YH, Ko YH. Benign and malignant mucocele-like tumors of the breast: mammographic and sonographic appearances. AJR 2005; 185:1310–1316 15. Kim SM, Kim HH, Kang DK, et al. Mucocele-like tumors of the breast as cystic lesions: sonographicpathologic correlation. AJR 2011; 196:1424–1430 16. Cardenosa G, Doudna C, Eklund GW. Mucinous (colloid) breast cancer: clinical and mammographic findings in 10 patients. AJR 1994; 162:1077–1079
F O R YO U R I N F O R M AT I O N
This article is available for CME and Self-Assessment (SA-CME) credit that satisfies Part II requirements for maintenance of certification (MOC). To access the examination for this article, follow the prompts associated with the online version of the article.
210
AJR:204, January 2015