Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, cont

Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights. Risk of Upgrade of Atypical Ductal Hyperplasia after Stereotactic Breast Biopsy: Effects of Number of Foci and Complete Removal of Calcifi cations 1 ORIGINAL RESEARCH n BREAST IMAGING Jennifer R. Kohr, MD Peter R. Eby, MD Kimberly H. Allison, MD Wendy B. DeMartini, MD Robert L. Gutierrez, MD Sue Peacock, MS Constance D. Lehman, MD, PhD Purpose: Materials and Methods: To determine if patients with fewer than three foci of atypical ductal hyperplasia (ADH) who have all of their calcifications removed after stereotactic 9- or 11-gauge vacuum-assisted breast biopsy (VABB) have a rate of upgrade to malignancy that is sufficiently low to obviate surgical excision. An institutional review board approved, HIPAA-compliant retrospective review of 991 cases of consecutive 9- or 11-gauge stereotactic VABB performed during a 65-month period revealed 147 cases of atypia. One pathologist performed a blinded review of the results of procedures performed to assess for calcifications and confirmed ADH in 101 cases with subsequent surgical excision. Each large duct or terminal duct lobular unit containing ADH was considered a focus and counted. Postbiopsy mammograms were reviewed to determine whether all calcifications were removed. Upgrade to malignancy was determined from excisional biopsy pathology reports. Upgrade rates as a function of both number of foci and presence or absence of residual calcifications were calculated and compared by using x 2 tests. Results: Upgrade to malignancy occurred in 20 (19.8%) of the 101 cases. The upgrade rate was significantly higher in cases of three or more foci of ADH (15 [28%] of 53 cases) than in cases of fewer than three foci (five [10%] of 48 cases) ( P =.02). Upgrade rates were similar, regardless of whether all mammographic calcifications were removed (seven [17%] of 41 cases) or all were not removed (nine [20%] of 45 cases) ( P =.77). Upgrade occurred in two (12%) of 17 cases in which there were fewer than three ADH foci and all calcifications were removed. 1 From the Departments of Radiology (J.R.K., P.R.E., W.B.D., R.L.G., S.P., C.D.L.) and Pathology (K.H.A.), University of Washington Medical Center, Seattle, Wash; and Seattle Cancer Care Alliance, 825 Eastlake Ave E, G3-200, Seattle, WA 98109-1023. Received August 5, 2009; revision requested August 31; revision received September 25; accepted October 6; fi nal version accepted November 23. Supported by a Radiology Health Services Research Pilot grant from the University of Washington. Address correspondence to P.R.E. (e-mail: preby@u.washington.edu ). Conclusion: The upgrade rate is significantly higher when ADH involves at least three foci. Surgical excision is recommended even when ADH involves fewer than three foci and all mammographic calcifications have been removed, because the upgrade rate is 12%. q RSNA, 2010 q RSNA, 2010 Radiology: Volume 255: Number 3 June 2010 n radiology.rsna.org 723

Atypical ductal hyperplasia (ADH) is a proliferative breast lesion that is considered to increase a woman s lifetime risk of developing breast carcinoma ( 1 ). It is defined histopathologically as a ductal proliferation with some but not all of the features of ductal carcinoma in situ (DCIS) ( 2,3 ). Some authors also include in this definition a proliferation possessing all of the features of low-grade DCIS but limited in total extent ( 2,3 ). These histopathologic criteria were initially developed on the basis of follow-up from surgically sampled open breast biopsies ( 1 ). Percutaneous core-needle biopsy and vacuum-assisted breast biopsy (VABB) enable the diagnosis of clinically suspicious or imaging suspicious breast lesions with sufficient accuracy to avoid excisional biopsy in most cases ( 4 7 ). However, because of the potential for undersampling of DCIS or invasive cancers associated with neighboring ADH when 11- or 14-gauge needle cores are used, excisional biopsy is currently recommended when ADH is identified after core-needle biopsy or VABB ( 4 31 ). Upgrade occurs whenever a malignant lesion is found after additional tissue sampling for a highrisk lesion diagnosed at percutaneous biopsy. Specifically, in the context of this study, upgrade occurs when DCIS Advances in Knowledge n Upgrade to ductal carcinoma in situ or invasive carcinoma is significantly less likely when atypical ductal hyperplasia (ADH) involves fewer than three foci rather than three or more foci ( P =.02). n Upgrade rates are not signifi- cantly different, regardless of whether determinant mammographic calcifications are completely removed at stereotactic vacuum-assisted breast biopsy ( P =.6). n Upgrade of ADH still occurred in 12% of cases when there were fewer than three foci and all mammographic calcifications had been removed. or invasive cancer is found at excisional biopsy after ADH has been diagnosed as the highest-risk lesion at VABB or coreneedle biopsy. With improved tissue sampling and decreasing upgrade rates of ADH diagnosed at 11- and 14-gauge VABB, several study investigators have evaluated possible predictors of upgrade to define a subgroup of patients with ADH who might avoid surgical biopsy ( 8 10,13,14,16,21,32 37 ). Some have evaluated the characteristics of atypia, including extent of atypia, number of foci, and degree of atypia ( 8 10,13,14,21,32,35 37 ). A few studies have revealed groups of patients for example, those with ADH involving fewer than three foci and complete removal of the lesion at needle biopsy with no upgrade at surgery ( 33,36,37 ). That these authors have found variable upgrade rates in patients with ADH suggests that not all ADH lesions are the same and that there may be a spectrum analogous to low-, intermediate-, and high-grade DCIS or atypical lobular hyperplasia and lobular carcinoma in situ. Furthermore, some authors have found that upgrade to malignancy does not occur when fewer foci of ADH are identified at biopsy and/or when the lesion, either calcifications or a mass, is completely removed at biopsy ( 13,33,36,37 ). Others have found lower upgrade rates in similar groups of patients but conclude that upgrade rates remain too high to avoid excisional biopsy ( 9,10,32 ). As a result, there is conflicting evidence as to whether subgroups of patients with limited ADH and complete lesion removal at biopsy can safely avoid surgical excision. Our goal was to determine the upgrade rates of ADH on the basis of the number of foci involved and the com- Implication for Patient Care n There are no patient subpopula- tions with a less than 2% risk of missed malignancy who can safely avoid surgical excisional biopsy when ADH is diagnosed as the highest risk lesion at stereotactic breast biopsy. plete removal of mammographic calcifications with 11- and 9-gauge VABB. Specifically, we sought to ascertain if the upgrade rate in a group of patients with ADH involving fewer than three foci and all calcifications removed was sufficiently low to obviate surgical excision. Materials and Methods Our institutional review board approved this retrospective Health Information Portability and Accountability Act compliant study. The methods used to identify, in our database, patients who received a diagnosis of ADH by means of VABB have been reported previously ( 38,39 ). We reviewed our database for all cases of stereotactic VABB performed for calcifications between February 1, 2001, and June 30, 2006. Patients were referred for stereotactic VABB after a complete diagnostic workup for a mammographic abnormality that resulted in a classification of Breast Imaging Reporting and Data System (BI-RADS) category 4 (suspicious) or 5 (highly suggestive of malignancy) at final assessment and a recommendation for biopsy. Patients with a current diagnosis of breast cancer (BI-RADS category 6) were excluded. Imaging evaluation was performed with screen-film mammography from February 2001 through April 2004 and Published online before print 10.1148/radiol.09091406 Radiology 2010; 255:723 730 Abbreviations: ADH = atypical ductal hyperplasia BI-RADS = Breast Imaging Reporting and Data System DCIS = ductal carcinoma in situ VABB = vacuum-assisted breast biopsy Author contributions: Guarantors of integrity of entire study, J.R.K., P.R.E.; study concepts/study design or data acquisition or data analysis/ interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript fi nal version approval, all authors; literature research, J.R.K., P.R.E., K.H.A., C.D.L.; clinical studies, J.R.K., K.H.A., C.D.L.; statistical analysis, P.R.E., S.P.; and manuscript editing, all authors Authors stated no fi nancial relationship to disclose. 724 radiology.rsna.org n Radiology: Volume 255: Number 3 June 2010

with full-field digital mammography from April 2004 through June 2006. Stereotactic VABB procedures were performed with the patient prone on a dedicated biopsy table (Lorad; Hologic, Bedford, Mass). From February 2001 through November 2003, 391 consecutive procedures were performed with an 11-gauge directional VABB device (Mammotome; Ethicon Endo-Surgery, Cincinnati, Ohio). From December 2003 through June 2006, 600 consecutive procedures were performed with a 9-gauge directional VABB device (ATEC; Suros Surgical Systems, Indianapolis, Ind). Thus, there were a total of 991 biopsies. We searched our radiology information system database for all cases that were assigned a diagnosis of ADH or atypia by one of four pathologists during the study period. One hundred forty-seven cases of ADH were identified. The original VABB histopathologic specimen slides for the 147 cases of atypia were retrieved. During the period of study, radiographs of all biopsy specimens were obtained and assessed for the presence of calcifications. All needle core samples were submitted for histopathologic evaluation. Three hematoxylineosin stained levels of the specimen were automatically prepared. If calcifications were not identified in any of the initial three hematoxylin-eosin stained specimen levels or the original pathologist believed it was of diagnostic benefit for other reasons, additional deeperlevel specimens were obtained by the original pathologist. An additional 13 (or approximately 10% of the study cases) benign cases without atypia were added to the study cohort for the purposes of blinding the reviewing pathologist to the original diagnosis. All of the original VABB histopathologic specimen slides for all 160 cases, both with and without atypia, were then pooled and re-reviewed by a breast pathologist (K.H.A.) with 8 years of experience. The pathologist was blinded to the details of all previous medical records and results when she rendered the new diagnosis. Cases that contained any DCIS or invasive cancer, as determined by the reviewing pathologist, were excluded. Cases containing solely lobular in situ neoplasia were also excluded. An additional eight cases that were assessed solely for mass or asymmetry were excluded. However, cases of procedures that were performed primarily for suspicious calcifications but also involved masses were included. As defined by Ely and colleagues ( 37 ), each large duct or terminal duct lobular unit involved with ADH was defined as one focus of ADH. The reviewing pathologist counted and recorded the total number of foci for all confirmed cases of ADH, applying the same diagnostic criteria established by Page and Rogers ( 3 ) and Tavassoli and Norris ( 2 ). The presence or absence of calcifications in association with the ADH was also recorded. Calcifications associated with ADH were considered to be determinant calcifications, whereas those that were not associated with ADH were considered unrelated calcifications. Only those cases in which the VABB target contained calcifications, either determinant or unrelated, and was assessed at surgical follow-up were included in the final upgrade analysis. The original diagnostic and immediate postbiopsy mammograms were retrieved and reviewed to determine the presence or absence of residual calcifications immediately after biopsy. The mammograms were reviewed in conjunction with the available radiology reports. For each case, the mammograms, either screen-film or digital, were reviewed by one radiology resident (J.R.K.), who was blinded to the final histopathology outcomes. The location of the targeted calcifications was confirmed from the biopsy report. The calcifications were then judged to be either completely or incompletely removed. In the event that there was more than one group of calcifications in the same region and only one group was targeted for biopsy, only the presence or absence of the targeted calcifications was evaluated. If the reviewing resident had any uncertainty regarding the presence or absence of residual calcification, the images were reviewed by a fellowship-trained breast imager with 5 years of experience (P.R.E.). If the pre- or postbiopsy mammograms could not be retrieved, then the status of calcification removal was defined as unknown. Upgrade to malignancy was determined by means of electronic medical record review on the basis of the final histopathologic diagnosis cited in the excisional biopsy reports. Upgrade as a function of both number of foci and presence or absence of residual mammographic calcifications, including determinant and unrelated calcifications, was calculated and compared with x 2 tests. For the purposes of analysis, the upgrade rate was also dichotomized into cases with three or more foci and those with fewer than three foci. This specific dichotomy was chosen because others have divided cases into these same two groups ( 32,33,36 ). Statistical software (SAS, version 9.1.3; SAS Institute, Cary, NC) was used, and P,.05 was considered to indicate significance. Results The breast pathologist s (K.H.A.) review of the 160 cases selected from the database of 991 lesions sampled with 9- or 11-gauge stereotactic VABB during the study period confirmed ADH without DCIS or invasive carcinoma in 112 cases. Surgical follow-up data were available for 101 (90%) of the 112 lesions. Upgrade to DCIS or invasive carcinoma occurred in 20 (19.8%) of these 101 cases (17 cases upgraded to DCIS, two upgraded to invasive carcinoma, and one upgraded to both DCIS and invasive carcinoma). The upgrade rates are summarized in Table 1. ADH was present in one focus (24.8%) in 25 of the 101 cases, in two foci in 23 (22.8%) cases, in three foci in 15 (14.9%) cases, and in four or more foci in 38 (37.6%) cases ( Table 2 ). Upgrade occurred in three of the 25 cases with one focus (12%), in two of the 23 cases with two foci (8.7%), in two of the 15 cases with three foci (13%), and in 13 of the 38 cases with four or more foci (34%). Upgrade was significantly more likely in the cases with at least three foci (15 [28%] of 53 cases) than in the cases with fewer than three foci (five [10%] of 48 cases) ( P =.02) ( Table 1 ). Postbiopsy mammograms were available for review in 86 of the 101 cases. Radiology: Volume 255: Number 3 June 2010 n radiology.rsna.org 725

Table 1 ADH Upgrade after Surgical Excision, Based on Histopathologic Criteria and Presence or Absence of Residual Calcifications Parameter Upgrade 95% Confi dence Interval * Histopathologic fi ndings All ADH 20/101 (19.8) 12, 27.6, 3 Foci of ADH 5/48 (10) 1.8, 19.1 3 Foci of ADH 15/53 (28) 16.2, 40.4 P value.02 Postbiopsy mammographic fi ndings Cases without residual calcifi cations 7/41 (17) 5.6, 28.6 Cases with residual calcifi cations 9/45 (20) 8.3, 31.7 P value.77 Determinant calcifi cations Cases without residual determinant calcifi cations 5/29 (17) 3.5, 30.6 Cases with residual determinant calcifi cations 8/36 (22) 8.6, 35.8 P value.6 No residual calcifi cations and, 3 ADH foci 2/17 (12) 0, 27.1 Note. Numbers in parentheses are percentages. * Data are 95% confi dence intervals of the percentages (in parentheses) in the Upgrade column. One case involving two ADH foci was upgraded to DCIS after complete removal of all determinant calcifi cations at 9-gauge VABB, and one case involving one ADH was upgraded to intraductal carcinoma when unrelated calcifi cations were removed at 11-gauge VABB. Calcifications were completely removed in 41 (48%) of the 86 cases, with the remaining 45 cases (52%) demonstrating incomplete removal of the targeted calcifications ( Table 1 ). The upgrade rate associated with the removal of all mammographic calcifications (17%, seven of 41 cases) was not significantly different from that associated with residual calcifications (20%, nine of 45 cases) ( P =.77). Of the 41 cases in which calcifications were completely removed after biopsy, 29 involved determinant calcifications at re-review of the histopathologic findings. Of the 45 cases with residual calcifications after biopsy, 36 were those with determinant calcifications. The upgrade rate associated with the removal of all determinant mammographic calcifications (17%, five of 29 cases) was not significantly different from that associated with residual determinant calcifications (22%, eight of 36 cases) ( P =.6) ( Table 1 ). There were 17 cases with fewer than three foci of ADH and all calcifications removed with 9-gauge ( n = 11) or 11-gauge ( n = 6) VABB. Upgrade occurred in two (12%) of these 17 cases. Ten of these 17 lesions contained deter- minant calcifications. One case had a single focus of ADH at biopsy, and after complete removal of the unrelated calcifications with 11-gauge VABB, it was subsequently upgraded to invasive ductal carcinoma at surgery. The second case involved two foci of ADH, complete removal of all determinant calcifications with 9-gauge VABB, and upgrade to DCIS ( Table 1 ). Discussion The accuracy of stereotactic core-needle biopsy of suspicious breast lesions has been well established for benign and malignant disease, greatly reducing the need for surgical excisional biopsy. Previous research has documented a general trend toward reduced ADH upgrade rates, initially with increasing needle gauge and subsequently with the advent of vacuum-assisted devices ( 17,20 22,28,40 ). Despite this, overall upgrade rates have remained too high to safely avoid the currently recommended surgical excisional biopsy. Specifically, upgrade rates remain greater than 2% the risk of malignancy that is widely accepted for the mammographic BI-RADS 3 category of probably benign and shortterm imaging follow-up as an alternative to tissue sampling. Some investigators have attempted to define a subpopulation of patients with a diagnosis of ADH who may be able to safely forgo surgical excision on the basis of certain histopathologic and radiographic features ( Tables 3 and 4 ) ( 8 10,13,14,16,32 34,36,37 ). We reviewed both the histopathologic and radiographic features of ADH diagnosed with 9- and 11-gauge VABB to further explore this possibility. Upgrade to malignancy occurred in five (10%) of 48 lesions with fewer than three foci of ADH. We observed an upgrade rate of 12% (two of 17 cases) when ADH was present in fewer than three ducts and all calcifications were removed. This occurred with both 11-gauge (17%, one of 16 cases) and 9-gauge (9%, one of 11 cases) VABB devices. In addition, we found no significant difference in upgrade rates between cases with removal of all determinant mammographic calcifications (in five [17%] of 29 cases) and cases with removal of residual determinant calcifications (in eight [22%] of 36 cases) ( P =.6). These results differ from those previously reported. Ely et al ( 37 ) found no upgrades among 24 lesions with fewer than three foci of ADH sampled at biopsy with 11- and 14-gauge biopsy devices. Sneige et al ( 36 ) used 11- and 14-gauge VABB needles and found no upgrade in a similar group (0 of 28 cases). In our study, one pathologist reviewed all the cases of ADH or atypia, confirmed the diagnosis of ADH, and, by using the same histopathologic criteria described by Ely et al ( 37 ) and used by Sneige et al ( 36 ), recorded the number of foci involved with ADH. Although we observed a significant difference between upgrade rates when ADH involved fewer than three foci rather than three or more foci (10% and 28%, respectively; P =.02), an upgrade rate of 10% is still too high to suggest that surgical excisional biopsy can be safely avoided on the basis of the number of foci of ADH alone. Although our conclusion is in contradiction to that of Sneige et al and Ely et al, the trend toward reduced 726 radiology.rsna.org n Radiology: Volume 255: Number 3 June 2010

Table 2 Cases of ADH with Surgical Follow-up Included in Upgrade Analysis and Sorted according to Foci, Needle Gauge, and Residual Calcifications Parameter 9-gauge Needle 11-gauge Needle All Cases All cases 62 (13) 39 (7) 101 (20) Calcifi cations remaining 30 (6) 15 (3) 45 (9), 3 ADH foci 18 (2) 5 (1) 23 (3) One focus 7 (1) 4 (1) 11 (2) Two foci 11 (1) 1 (0) 12 (1) 3 ADH foci 12 (4) 10 (2) 22 (6) Three foci 4 (1) 4 (1) 8 (2). 3 Foci 8 (3) 6 (1) 14 (4) Calcifi cations removed 25 (4) 16 (3) 41 (7), 3 ADH foci 11 (1) 6 (1) 17 (2) One focus 5 (0) 4 (1)* 9 (1) Two foci 6 (1)* 2 (0) 8 (1) 3 ADH foci 14 (3) 10 (2) 24 (5) Three foci 5 (0) 0 (0) 5 (0). 3 Foci 9 (3) 10 (2) 19 (5) Calcifi cation status unknown 7 (3) 8 (1) 15 (4), 3 ADH foci 2 (0) 6 (0) 8 (0) One focus 1 (0) 4 (0) 5 (0) Two foci 1 (0) 2 (0) 3 (0) 3 ADH foci 5 (3) 2 (1) 7 (4) Three foci 2 (0) 0 (0) 2 (0). 3 Foci 3 (3) 2 (1) 5 (4) Note. Data are numbers of cases. Numbers in parentheses are numbers of upgrades. * Cases include two cases of upgrade with fewer than three foci of ADH and all calcifi cations removed. Table 3 Published Data on ADH Upgrade versus Number of Foci Study Biopsy Method Overall Upgrade Rate upgrade with less-extensive ADH is consistent with their findings and with the findings of others ( 9,32,33 ) ( Table 3 ). One possible explanation for the disparity is that our current study sample size of ADH with fewer than three foci was Upgrade Rate with,3 ADH Foci Upgrade Rate with 3 ADH Foci Bonnett et al ( 9 )* 11-Gauge VABB and 30/73 (41) 16/49 (33) 14/24 (58) 14-gauge CNB Ely et al ( 37 ) 11- and 14-Gauge 17/47 (36) 0/24 (0) 17/23 (74) Forgeard et al ( 32 ) 11-Gauge VABB 29/116 (25) 5/44 (11) 24/72 (33) Sneige et al ( 36 ) 11- and 14-Gauge VABB 3/42 (7.1) 0/28 (0) 3/14 (21) Wagoner et al ( 33 ) 11-Gauge VABB and 22/123 (17.9) 6/82 (7.3) 16/41 (39) 14-gauge CNB Current study 9- and 11-Gauge VABB 20/101 (19.8) 5/48 (10) 15/53 (28) Note. Except for the Study column, numbers in parentheses are percentages. CNB = core-needle biopsy. * This study included ADH, atypical lobular hyperplasia, and a heterogeneous group of non-adh or atypical lobular hyperplasia. VABB or core-needle biopsy was not specifi ed in this study. approximately double that in either of these studies (48 lesions vs 24 lesions for Ely et al and 28 lesions for Sneige et al). In addition, the overall upgrade rate for Sneige et al (7%) was among the lowest reported. The upgrade rates and sample sizes of other studies were more similar to ours ( 32,33 ). Sneige et al ( 36 ), Jackman et al ( 10 ), Wagoner et al ( 33 ), and others also examined the rate of ADH upgrade when percutaneous biopsy resulted in the complete removal of all suspicious calcifications or the entire mass ( 32,37,41 ). Sneige et al ( 36 ) observed no upgraded lesion when all calcifications were removed (0 of 15 cases). Although Jackman et al ( 10 ) observed a significant difference in the rates of upgrade when either all calcifications or the entire mass was removed at biopsy (8% [three of 36 cases] vs 29% [19 of 66 cases], respectively), the upgrade rates remained too high to avoid surgical excision despite the presumed increased sample size with 11-gauge VABB needles. Similarly, Wagoner et al ( 33 ) observed an upgrade rate of 5% (three of 57 cases) when all calcifications were removed versus a rate of 24% (19 of 79 cases) when there were residual calcifications. When all determinant calcifications (those associated with ADH histopathologically) were removed, however, there was a 3% (one of 34 cases) upgrade rate versus a 35% (14 of 40 cases) upgrade rate when there were residual determinant calcifications. Finally, there were no upgrades in the group with all determinant calcifications removed and ADH involving fewer than three foci (0 of 25 cases). Interestingly, both Wagoner et al ( 33 ) and Jackman et al ( 10 ) used 11-gauge VABB needles exclusively (for VABB of calcifications for Wagoner et al), whereas Sneige et al ( 36 ) used both 11- and 14-gauge needles. We observed no difference in the upgrade rate when suspicious calcifications were completely removed at VABB (seven of 41 cases, 17%) versus when they were not (nine of 45 cases, 20%), and the same was true with respect to determinant calcifications (five [17%] of 29 cases vs eight [22%] of 36 cases, respectively). Similar to Sneige et al ( 36 ) and Wagoner et al ( 33 ), we evaluated a group of lesions with fewer than three foci of ADH and all calcifications removed and observed a 12% (two of 17 cases) upgrade rate. This was essentially no different from the upgrade rate of lesions Radiology: Volume 255: Number 3 June 2010 n radiology.rsna.org 727

Table 4 Published Data on ADH Upgrade as Function of Complete versus Incomplete Lesion Removal at Image-guided Needle Biopsy Study Biopsy Device Overall Upgrade Rate Upgrade Rate When Lesion Completely Removed Upgrade Rate When Lesion Incompletely Removed Studies including any lesion type Jackman et al ( 10 ) 11-Gauge VABB 22/104 (21)* 3/36 (8) 19/66 (29) Philpotts et al ( 14 ) 11-Gauge VABB 14/94 (15) 5/40 (12) 9/54 (17) Liberman et al ( 41 ) 11-Gauge VABB 12/49 (24) 6/32 (19) 5/16 (31) Studies including biopsies for calcifi cations Adrales et al ( 13 ) 11-Gauge VABB 9/62 (14) 0/30 (0) 9/32 (28) Forgeard et al ( 32 ) 11-Gauge VABB 29/116 (25) 10/50 (20) 19/66 (29) Sneige et al ( 36 ) 11- and 14-Gauge VABB 3/42 (7) 0/15 (0) 3/27 (11) Wagoner et al ( 33 ) 11-Gauge VABB 20/110 (18.2) 3/57 (5.2) 19/79 (24) Current study 9- and 11-Gauge VABB 20/101 (19.8) 7/41 (17) 9/45 (20) Note. Except for the Study column, numbers in parentheses are percentages. * Two lesions were not examined at postbiopsy imaging. Lesions were calcifi cations and architectural distortions. ADH and DCIS lesions were combined before being divided into 75% or less removed and greater than 75% removed groups. Studies included only biopsies performed primarily for calcifi cations. All types of suspicious abnormalities were evaluated, but only calcifi cations (both determinant and unrelated) are included in the presented data. with fewer than three foci of ADH (five [10%] of 48 lesions) or the upgrade rate observed when the lesion contained fewer than three foci of ADH and all determinant calcifications were removed (one [10%] of 10 cases). The combination of these findings suggests that complete excision of the mammographic lesion by means of VABB does not sufficiently decrease the upgrade rate to safely avoid surgical excision. This conclusion contradicts those of Sneige et al ( 36 ) and Wagoner et al ( 33 ), both of whom identified subgroups with no upgrade who, they suggested, may have been able to avoid surgical excision. Whether a subgroup has a low enough upgrade rate to avoid surgical biopsy has important implications for patient treatment. The additional cost, risk, and breast deformity inherent with excisional biopsy must be weighed against the risk of nondiagnosed malignancy. Ideally, those patients with a truly increased risk of malignancy should be referred for excisional biopsy. The current data, including our own, suggest that ADH is a spectrum with varying risk of associated malignancy. However, until further consistent stratification can be performed histopathologically and radiographically, all biopsied lesions that yield ADH should be excised. Although the BI-RADS 3 category of probably benign cannot be strictly applied in this clinical scenario, reducing the risk of nondiagnosed malignancy to less than 2% in a subgroup of patients with ADH defined histopathologically and radiographically would yield an upgrade rate at which mammographic follow-up would be acceptable. We previously published data showing that there was no marked difference in upgrade rates between 9- and 11-gauge VABB groups ( 39 ). Others have found similar results, with no significant decrease in upgrade rates with increasing needle gauge ( 40 ). Furthermore, the results of a study by Liberman et al ( 41 ) suggest that complete excision of breast lesions confers no substantial advantage in terms of reduced risk of imaginghistopathologic discordance, missed lesions, upgrade rate, repeat biopsy, or surgery across benign, high-risk, and cancerous breast lesions. Liberman et al also suggest that the mammogram itself may underestimate the extent of the lesion biopsied. Although more research must be done to confirm this, these early comparisons suggest that there may be no further benefit to increasing the needle and/or sample size. Perhaps the focus should be on finding consistent variables for risk stratification of ADH with the goal of referring only those patients with a truly increased risk of malignancy for surgical excision. One limitation of this study was that only a subset of histopathologic slides were reviewed rather than slides of all 991 specimens from the study period, the assessment of which would have been prohibitively time consuming. It is possible that a few cases of ADH or atypia were not included in the study cohort. However, it is unlikely that these affected the results. In addition, the purpose of the review was to apply uniform histopathologic criteria and count the number of foci of ADH not to validate these criteria in the initial study group of 991 lesions. Another minor limitation was that a radiology resident reviewed the mammograms and independently made the final determination in most cases. Because this evaluation was relatively simple, regardless of the presence of residual calcifications, and those cases in which the resident had questions were reviewed by an experienced breast imager (P.R.E.), we believe that the data are adequate. In summary, we found that upgrade to DCIS or invasive carcinoma was significantly less likely when ADH involved fewer than three foci than when it involved three or more foci ( P =.02). In 728 radiology.rsna.org n Radiology: Volume 255: Number 3 June 2010

contrast, there were no significant differences in upgrade rates based on whether the determinant mammographic calcifications were completely removed at stereotactic VABB ( P =.6). Ultimately, despite the theoretically improved accuracy of 9- and 11-gauge VABB needles and the risk stratification performed with histopathologic and mammographic criteria, we were unable to identify a subpopulation of patients with ADH who could avoid surgical excision. Thus, upgrade of ADH still occurs when there are fewer than three foci and all mammographic calcifications have been removed. Therefore, surgical excision is recommended for all patients with a diagnosis of ADH after stereotactic VABB for calcifications. References 1. Page DL, Dupont WD, Rogers LW, Rados MS. Atypical hyperplastic lesions of the female breast: a long-term follow-up study. Cancer 1985 ; 55 ( 11 ): 2698 2708. 2. Tavassoli FA, Norris HJ. A comparison of the results of long-term follow-up for atypical intraductal hyperplasia and intraductal hyperplasia of the breast. Cancer 1990 ; 65 ( 3 ): 518 529. 3. Page DL, Rogers LW. Combined histologic and cytologic criteria for the diagnosis of mammary atypical ductal hyperplasia. Hum Pathol 1992 ; 23 ( 10 ): 1095 1097. 4. Jackman RJ, Nowels KW, Rodriguez-Soto J, Marzoni FA Jr, Finkelstein SI, Shepard MJ. Stereotactic, automated, large-core needle biopsy of nonpalpable breast lesions: false-negative and histologic underestimation rates after long-term follow-up. Radiology 1999 ; 210 ( 3 ): 799 805. 5. Fuhrman GM, Cederbom GJ, Bolton JS, et al. Image-guided core-needle breast biopsy is an accurate technique to evaluate patients with nonpalpable imaging abnormalities. Ann Surg 1998 ; 227 ( 6 ): 932 939. 6. Lee CH, Egglin TK, Philpotts L, Mainiero MB, Tocino I. Cost-effectiveness of stereotactic core needle biopsy: analysis by means of mammographic findings. Radiology 1997 ; 202 ( 3 ): 849 854. 7. Jackman RJ, Nowels KW, Shepard MJ, Finkelstein SI, Marzoni FA Jr. Stereotaxic large-core needle biopsy of 450 nonpalpable breast lesions with surgical correlation in lesions with cancer or atypical hyperplasia. Radiology 1994 ; 193 ( 1 ): 91 95. 8. Yeh IT, Dimitrov D, Otto P, Miller AR, Kahlenberg MS, Cruz A. Pathologic review of atypical hyperplasia identified by imageguided breast needle core biopsy: correlation with excision specimen. Arch Pathol Lab Med 2003 ; 127 ( 1 ): 49 54. 9. Bonnett M, Wallis T, Rossmann M, et al. Histopathologic analysis of atypical lesions in image-guided core breast biopsies. Mod Pathol 2003 ; 16 ( 2 ): 154 160. 10. Jackman RJ, Birdwell RL, Ikeda DM. Atypical ductal hyperplasia: can some lesions be defined as probably benign after stereotactic 11-gauge vacuum-assisted biopsy, eliminating the recommendation for surgical excision? Radiology 2002 ; 224 ( 2 ): 548 554. 11. Lai JT, Burrowes P, MacGregor JH. Diagnostic accuracy of a stereotaxically guided vacuum-assisted large-core breast biopsy program in Canada. Can Assoc Radiol J 2001 ; 52 ( 4 ): 223 227. 12. Cangiarella J, Waisman J, Symmans WF, et al. Mammotome core biopsy for mammary microcalcification: analysis of 160 biopsies from 142 women with surgical and radiologic followup. Cancer 2001 ; 91 ( 1 ): 173 177. 13. Adrales G, Turk P, Wallace T, Bird R, Norton HJ, Greene F. Is surgical excision necessary for atypical ductal hyperplasia of the breast diagnosed by Mammotome? Am J Surg 2000 ; 180 ( 4 ): 313 315. 14. Philpotts LE, Lee CH, Horvath LJ, Lange RC, Carter D, Tocino I. Underestimation of breast cancer with 11-gauge vacuum suction biopsy. AJR Am J Roentgenol 2000 ; 175 ( 4 ): 1047 1050. 15. Burak WE Jr, Owens KE, Tighe MB, et al. Vacuum-assisted stereotactic breast biopsy: histologic underestimation of malignant lesions. Arch Surg 2000 ; 135 ( 6 ): 700 703. 16. Liberman L, Smolkin JH, Dershaw DD, Morris EA, Abramson AF, Rosen PP. Calcification retrieval at stereotactic, 11-gauge, directional, vacuum-assisted breast biopsy. Radiology 1998 ; 208 ( 1 ): 251 260. 17. Jackman RJ, Burbank F, Parker SH, et al. Atypical ductal hyperplasia diagnosed at stereotactic breast biopsy: improved reliability with 14-gauge, directional, vacuum-assisted biopsy. Radiology 1997 ; 204 ( 2 ): 485 488. 18. Berg WA, Krebs TL, Campassi C, Magder LS, Sun CC. Evaluation of 14- and 11-gauge directional, vacuum-assisted biopsy probes and 14-gauge biopsy guns in a breast parenchymal model. Radiology 1997 ; 205 ( 1 ): 203 208. 19. Burbank F. Specimen weights obtained with 14- and 11-gauge breast biopsy probes. Radiology 1998 ; 208 ( 1 ): 269 270. 20. Darling ML, Smith DN, Lester SC, et al. Atypical ductal hyperplasia and ductal carcinoma in situ as revealed by largecore needle breast biopsy: results of surgical excision. AJR Am J Roentgenol 2000 ; 175 ( 5 ): 1341 1346. 21. O hea BJ, Tornos C. Mild ductal atypia after large-core needle biopsy of the breast: is surgical excision always necessary? Surgery 2000 ; 128 ( 4 ): 738 743. 22. Meyer JE, Smith DN, Lester SC, et al. Large-core needle biopsy of nonpalpable breast lesions. JAMA 1999 ; 281 ( 17 ): 1638 1641. 23. Lin PH, Clyde JC, Bates DM, Garcia JM, Matsumoto GH, Girvin GW. Accuracy of stereotactic core-needle breast biopsy in atypical ductal hyperplasia. Am J Surg 1998 ; 175 ( 5 ): 380 382. 24. Brown TA, Wall JW, Christensen ED, et al. Atypical hyperplasia in the era of stereotactic core needle biopsy. J Surg Oncol 1998 ; 67 ( 3 ): 168 173. 25. Gadzala DE, Cederbom GJ, Bolton JS, et al. Appropriate management of atypical ductal hyperplasia diagnosed by stereotactic core needle breast biopsy. Ann Surg Oncol 1997 ; 4 ( 4 ): 283 286. 26. Moore MM, Hargett CW III, Hanks JB, et al. Association of breast cancer with the finding of atypical ductal hyperplasia at core breast biopsy. Ann Surg 1997 ; 225 ( 6 ): 726 733. 27. Liberman L, Dershaw DD, Glassman JR, et al. Analysis of cancers not diagnosed at stereotactic core breast biopsy. Radiology 1997 ; 203 ( 1 ): 151 157. 28. Burbank F. Stereotactic breast biopsy of atypical ductal hyperplasia and ductal carcinoma in situ lesions: improved accuracy with directional, vacuum-assisted biopsy. Radiology 1997 ; 202 ( 3 ): 843 847. 29. Nguyen M, McCombs MM, Ghandehari S, et al. An update on core needle biopsy for radiologically detected breast lesions. Cancer 1996 ; 78 ( 11 ): 2340 2345. 30. Tocino I, Garcia BM, Carter D. Surgical biopsy findings in patients with atypical hyperplasia diagnosed by stereotaxic core needle biopsy. Ann Surg Oncol 1996 ; 3 ( 5 ): 483 488. 31. Liberman L, Cohen MA, Dershaw DD, Abramson AF, Hann LE, Rosen PP. Atypical ductal hyperplasia diagnosed at stereotaxic core biopsy of breast lesions: an indication Radiology: Volume 255: Number 3 June 2010 n radiology.rsna.org 729

for surgical biopsy. AJR Am J Roentgenol 1995 ; 64 ( 5 ): 1111 1113. 32. Forgeard C, Benchaib M, Guerin N, et al. Is surgical biopsy mandatory in case of atypical ductal hyperplasia on 11- gauge core needle biopsy? a retrospective study of 300 patients. Am J Surg 2008 ; 196 ( 3 ): 339 345. 33. Wagoner MJ, Laronga C, Acs G. Extent and histologic pattern of atypical ductal hyperplasia present on core needle biopsy specimens of the breast can predict ductal carcinoma in situ in subsequent excision. Am J Clin Pathol 2009 ; 131 ( 1 ): 112 121. 34. Renshaw AA, Cartagena N, Schenkman RH, Derhagopian RP, Gould EW. Atypical ductal hyperplasia in breast core needle biopsies: correlation of size of the lesion, complete removal of the lesion, and the incidence of carcinoma in follow-up biopsies. Am J Clin Pathol 2001 ; 116 ( 1 ): 92 96. 35. Doren E, Hulvat M, Norton J, et al. Predicting cancer on excision of atypical ductal hyperplasia. Am J Surg 2008 ; 195 ( 3 ): 358 362. 36. Sneige N, Lim SC, Whitman GJ, et al. Atypical ductal hyperplasia diagnosis by directional vacuum-assisted stereotactic biopsy of breast microcalcifications: considerations for surgical excision. Am J Clin Pathol 2003 ; 119 ( 2 ): 248 253. 37. Ely KA, Carter BA, Jensen RA, Simpson JF, Page DL. Core biopsy of the breast with atypical ductal hyperplasia: a probabilistic approach to reporting. Am J Surg Pathol 2001 ; 25 ( 8 ): 1017 1021. 38. Eby PR, Ochsner JE, DeMartini WB, Allison KH, Peacock S, Lehman CD. Is surgical excision necessary for focal atypical ductal hyperplasia found at stereotactic vacuum-assisted breast biopsy? Ann Surg Oncol 2008 ; 15 ( 11 ): 3232 3238. 39. Eby PR, Ochsner JE, DeMartini WB, Allison KH, Peacock S, Lehman CD. Frequency and upgrade rates of atypical ductal hyperplasia diagnosed at stereotactic vacuum-assisted breast biopsy: 9- versus 11-gauge. AJR Am J Roentgenol 2009 ; 192 ( 1 ): 229 234. 40. Lourenco AP, Mainiero MB, Lazarus E, Giri D, Schepps B. Stereotactic breast biopsy: comparison of histologic underestimation rates with 11- and 9-gauge vacuum-assisted breast biopsy. AJR Am J Roentgenol 2007 ; 189 ( 5 ): W275 W279. 41. Liberman L, Kaplan JB, Morris EA, Abramson AF, Menell JH, Dershaw DD. To excise or to sample the mammographic target: what is the goal of stereotactic 11-gauge vacuumassisted breast biopsy? AJR Am J Roentgenol 2002 ; 179 ( 3 ): 679 683. 730 radiology.rsna.org n Radiology: Volume 255: Number 3 June 2010