EARLY ONLINE RELEASE

Transcription

1 EARLY ONLINE RELEASE Note: This article was posted on the Archives Web site as an Early Online Release. Early Online Release articles have been peer reviewed, copyedited, and reviewed by the authors. Additional changes or corrections may appear in these articles when they appear in a future print issue of the Archives. Early Online Release articles are citable by using the Digital Object Identifier (DOI), a unique number given to every article. The DOI will typically appear at the end of the abstract. The DOI for this manuscript is doi: /arpa OA The final published version of this manuscript will replace the Early Online Release version at the above DOI once it is available College of American Pathologists

2 Original Article Triaging Atypical Lobular Hyperplasia and Lobular Carcinoma In Situ on Percutaneous Core Biopsy to Surgery or Observation Assiduous Radiologic-Pathologic Correlation Works, Quantitating Extent of Disease Does Not Anna I. Holbrook, MD; Krisztina Hanley, MD; Caprichia Jeffers, MS; Jian Kang, PhD; Michael A. Cohen, MD Context. The management of lobular neoplasia (LN) found on core biopsy is controversial and ranges from obligatory surgical excision to clinical/imaging observation. Objective. To determine if in cases of core needle biopsy yielding LN, quantification of normal and diseased terminal ductal lobular units (TDLUs) can predict which cases require surgical excision and which can be safely followed. A secondary goal is to assess whether the concordance or discordance of core biopsy results, determined by rigorous radiologic-pathologic correlation, can predict for upgrade to malignancy at excision. Design. In this retrospective study, 79 specimens from 78 women who underwent image-guided core needle biopsies between 2005 and 2012 yielding LN were evaluated for total number of TDLUs and total number and percentage of TDLUs involved by LN. Additionally, radiologic-pathologic correlation was performed to assess concordance or discordance. All were correlated with the results of surgical excisional biopsy or imaging/clinical follow-up. Results. There were 5 upgrades to malignancy. There was no association between upgrade to malignancy and any of the 3 TDLU variables evaluated, including total TDLUs in the specimen (P ¼.42), total abnormal TDLUs (P ¼.56), and percent of TDLUs that are abnormal (P ¼.07). Kendall rank correlation demonstrated a correlation between discordance and upgrade to cancer at surgery that was statistically significant (sb ¼ 0.394, P,.001). Conclusions. Quantifying total TDLU and those involved by LN on core biopsy will not aid in triaging patients to surgery or observation. Assiduous radiologicpathologic correlation to determine lesion concordance/ discordance can predict those patients who would benefit from surgical excision. (Arch Pathol Lab Med. doi: /arpa OA) The management of atypical lobular hyperplasia (ALH) surgical excision ranging from 8% to 34%, 1 15 others have and lobular carcinoma in situ (LCIS), together known demonstrated significantly lower upgrade rates in the range as lobular neoplasia (LN), as the most significant lesion on of 0% to 3% As a result of the conflicting data, core biopsy remains controversial in spite of a relatively management at various institutions and practices ranges large number of publications dealing with the topic. from obligatory surgical excision of either ALH or LCIS Although many articles have detailed high upgrade rates to ductal carcinoma in situ (DCIS) or invasive carcinoma at obtained on core biopsy to surgical excision in only specific limited instances with clinical and imaging observation for the remaining majority. Several investigators have reported various features of the Accepted for publication July 10, From the Departments of Radiology and Imaging Sciences (Drs pathologic core biopsy specimens that might assist in Holbrook and Cohen), Pathology and Laboratory Medicine (Dr predicting for upgrade, but again with mixed results. Hanley), and Biostatistics and Bioinformatics (Ms Jeffers), Rollins 20,21,24 26,28,29 We defined 3 reproducible and quantifi- School of Public Health, Emory University, Atlanta, Georgia; and the able features inherent in the core samples that might predict Department of Biostatistics, University of Michigan, Ann Arbor (Dr for upgrade: total number of terminal ductal lobular units Kang). The authors have no relevant financial interest in the products or (TDLUs) in the core specimens as a predictor of overall companies described in this article. sampling adequacy, total number of TDLUs containing ALH Presented in part at the American Roentgen Ray Society annual or LCIS as a predictor of overall disease extent in the core meeting; April 20, 2016; Los Angeles, California. specimens, and ratio of ALH/LCIS cores to total overall Corresponding author: Anna Holbrook, MD, Department of Radiology and Imaging Sciences, Emory University, 1365C Clifton number of TDLUs as an estimate of disease density within Road NE, Building C, Suite C1104, Atlanta, GA ( the total sampled specimen. The purpose of this study is to aiholbr@emory.edu). determine if in cases of core needle biopsy yielding LN, Arch Pathol Lab Med Quantifying TDLU Holbrook et al 1

3 quantification of normal and diseased TDLUs might assist in predicting which cases require surgical excision and which can be safely followed. Additionally, Atkins et al 21 concluded that in cases of ALH/LCIS found at core needle biopsy, careful radiologicpathologic correlation could be used to determine which patients should undergo surgical excision. They found that in 43 cases considered concordant, there were no malignancies found at surgery or follow-up, whereas there were 2 upgrades to DCIS among the 7 discordant biopsy cases. 21 A secondary goal of our study is to validate their results using similar methodology to assess whether the concordance or discordance of core biopsy results, determined by rigorous radiologic-pathologic correlation, can predict for upgrade to malignancy at excision. MATERIALS AND METHODS This retrospective study was Institutional Review Board approved with waiver of informed consent and was compliant with the Health Insurance Portability and Accountability Act. The pathology database at our institution was searched for consecutive cases in which an image-guided core needle biopsy performed between 2005 and 2012 yielded LCIS and/or ALH. Those in which there was another finding that would prompt excision (atypical ductal hyperplasia, flat epithelial atypia, papilloma, radial scar, DCIS, invasive carcinoma) were excluded and only those with available surgical excision results or at least 2 years of clinical and/or imaging follow-up were included. Other exclusion criteria included incomplete imaging and/or pathologic material. The pathologist (K.H., with 9 years of experience) reviewed the original slides from each core needle biopsy blinded to the outcome at surgical excision or extended follow-up. All submitted cores were embedded in paraffin blocks. On each tissue block, 3 levels (L1, L3, and L5), each containing 5-lm tissue sections, were examined. Total TDLU and ALH/LCIS-containing TDLU counts were performed on the level with the greatest number of TDLUs. If multiple tissue blocks were necessary to embed all of the cores, then the number of total TDLUs and ALH/LCIS-containing TDLUs was the sum of the TDLUs from the salient levels from each block examined. In the final TDLU count, both lobules and lobules with an associated terminal duct were counted. The number of TDLUs involved by ALH/LCIS was then quantitated. In cases where ALH/ LCIS involved a fibroadenoma, an attempt was made to count the TDLUs within and outside of the fibroadenoma. The images pertinent to the case, including the diagnostic mammogram, ultrasound, and magnetic resonance imaging (MRI), and images from the biopsy, were reviewed by 2 radiologists (with 35 [M.A.C.] and 7 [A.I.H.] years of experience) also blinded to the results of any subsequent surgical excision or extended imaging/ clinical follow-up. Concordance or discordance of the biopsy results with the radiologic findings was then determined by consensus by the radiologists and pathologist. If the histologic results could fully account for the imaging features, then the results were considered concordant. Otherwise, the results were designated as discordant. For example, if a mass was seen on imaging but no mass-forming entity was found at histology, this would be considered discordant. We also included in this category cases in which sampling was considered inadequate for example, ones in which there were few calcifications in the specimen radiograph. Lastly, imaging findings that are highly suggestive of malignancy and given Breast Imaging Reporting and Data System (BI-RADS) 5 designation would be called discordant because a benign core biopsy result in these cases would not be acceptable. Only after all cases were fully analyzed were the authors unblinded to the result of surgical excision or follow-up as determined by review of the medical record. Upgrade of any case to DCIS or invasive carcinoma was recorded. We modeled the probability of upgrade as a function of total abnormal TDLUs, total TDLUs, and percent abnormal TDLUs using univariate and multivariate logistic regression models. Kendall rank correlation was run to determine the relationship between concordance status and upgrade to cancer at surgery (R Core Team (2016). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL accessed July 2018). RESULTS There were 253 cases found in the pathology database in which an image-guided biopsy performed between 2005 and 2012 yielded LCIS and/or ALH. A total of 146 of these cases had a coexistent high-risk lesion or cancer in the specimen and were therefore excluded. Fifteen cases were lost to follow-up. Two patients were excluded because they received preoperative chemotherapy and radiation for treatment of synchronous cancer elsewhere in the breasts, and therefore upgrade at the biopsy site could not be assessed. There were 2 cases with missing images, and 9 cases were missing pathologic slides. One patient had 2 core needle biopsies in the left breast, 2 years apart, and at 2 separate sites. A total of 79 cases in 78 patients met the inclusion criteria. The patients ranged in age from 36 to 87 years (mean age, 58 years). The imaging finding that prompted the biopsy included mass (n ¼ 17, including 7 masses seen at MRI), architectural distortion (n ¼ 2), asymmetry (n ¼ 1), focal asymmetry (n ¼ 7), focus of enhancement on MRI (n ¼ 1), non mass-like enhancement on MRI (n ¼ 6), and calcifications (n ¼ 45). Calcifications were classified as pleomorphic (n ¼ 21), amorphous (n ¼ 16), and punctate (n ¼ 8). Cases were classified as BI-RADS 4 (n ¼ 75) or BI-RADS 5 (n ¼ 4). Biopsies were performed under stereotactic guidance in 47 cases, of which 45 cases had reports specifying needle gauge (43 cases with 11-gauge needle and 2 cases with 9-gauge needle). Of the 47 cases, 32 had reports specifying number of cores obtained (average, 11.1 cores; range, 6 15 cores; median, 12 cores). Biopsies were performed under ultrasound guidance in 21 cases. Of these, 17 had reports that specified the needle gauge used (1 case with 16-gauge needle, 14 cases with 14-gauge needle, and 2 cases with 11- gauge needle). Of the 21 cases, 15 had reports specifying the number of cores obtained (average, 4.8 cores, range, 4 8 cores; median, 4 cores). A total of 11 cases were obtained under MRI guidance with a 9-gauge needle, and of these, 7 had reports specifying the number of cores obtained (average, 12.1 cores; range, cores; median, 12 cores). Core needle biopsy specimens demonstrated ALH (n ¼ 52), LCIS (n ¼ 17), ALH and LCIS (n ¼ 5), LCIS with necrosis (n ¼ 1), and pleomorphic LCIS (n ¼ 4). Of the 45 cases in which calcifications prompted the biopsy, 11 demonstrated calcium intrinsically associated with the LN. The total number of TDLUs counted in each specimen ranged from 2 to 180 (mean, 47.4). The total number of TDLUs involved by ALH/LCIS in each specimen ranged from 1 to 38 (mean, 6.1). The percent of TDLUs involved by ALH/LCIS in each specimen ranged from 0.6% (1 of 180) to 100% (2 of 2; mean, 21%; Table 1). Following radiologic-pathologic review, core biopsy results were considered discordant in 24 of 79 cases (30%). The most common reason for discordance was a mass seen on imaging but no mass-forming entity found at histology (14 cases). In 5 cases, there was non mass-like enhancement on MRI that prompted the biopsy, but no sufficient explanation for the enhancement was seen at pathology. In 3 cases, the target of the biopsy was suspicious calcifications; 2 Arch Pathol Lab Med Quantifying TDLU Holbrook et al

4 Table 1. Quantified Terminal Ductal Lobular Unit (TDLUs) Total, Total Involved by Lobular Neoplasia (LN), and % Involved by LN and Association With Upgrade to Malignancy at Surgery Upgrade to Malignancy at Surgery? Total Sample, No. Total TDLU Total Abnormal TDLU % Abnormal TDLU Mean SD Mean SD Mean SD Yes No Summary P (univariate) P (multivariate) however, few calcifications were seen in the specimen radiograph. Lastly, in 2 cases the findings were given a BI- RADS 5 designation and were so highly suggestive of malignancy that a benign core biopsy result could not be accepted. A total of 66 cases then underwent surgical excision. In the remaining 13 cases, the patients did not undergo excision of the LN but had at least 2 years of imaging and/or clinical follow-up (average follow-up, 5.5 years; range, 2 years to 10 years, 9 months). There were a total of 5 upgrades to malignancy at surgical excision (Table 2). There were no cases of malignancy found during clinical/imaging follow-up. In all of the upgraded cases, the imaging features were deemed discordant with the pathology findings on core biopsy (Figure 1). There were no concordant cases that were upgraded to malignancy (Figure 2). Statistical analysis demonstrates that there was no association between upgrade to cancer at surgical excision and any of the 3 TDLU variables evaluated, including total TDLUs in the specimen (P ¼.42), total abnormal TDLUs (P ¼.56), and percent of TDLUs that were abnormal (P ¼.07). Multivariate analysis also showed no significant association between the 3 TDLU variables, evaluated jointly, and upgrade to malignancy at excision (total TDLU P ¼.93; total abnormal TDLU P ¼.95; percent abnormal TDLU P ¼.23; Figure 3; Table 1). Kendall rank correlation demonstrated a correlation between discordance and upgrade to cancer at surgery that was statistically significant (sb ¼ 0.394, P,.001). DISCUSSION Atkins et al 21 found that in the setting of core biopsy yielding ALH/LCIS, careful radiologic-pathologic correlation could determine which patients could safely avoid undergoing surgical excision. Our results support their conclusions, because we also found no upgrades among our cases determined concordant at radiologic-pathologic review. All 5 upgrades occurred only among the cases determined discordant at radiologic-pathologic review. Our data would strongly support the hypothesis that assiduous radiologic-pathologic review of core biopsy cases yielding lobular neoplasia on core biopsy can triage those requiring surgery from those that can be managed safely with clinical and imaging follow-up. Beginning with observations by Ottesen et al 30 in the early 1990s, there has been an interest in quantification of ALH/LCIS at biopsy as a predictor of more serious adjacent invasive or in situ disease. Although their work predated widespread percutaneous core biopsy, these authors found that greater than 10 lobules containing LCIS on surgical excisional biopsy predicted for future development of invasive tumor. 30 Subsequently, several publications quantifying ALH and LCIS on core biopsy have produced conflicting data as to whether the extent of ALH or LCIS found at core needle biopsy predicts for upgrade to malignancy at excision (summarized in Table 3). Several authors suggested that quantity of LN on core biopsy did predict for upgrade to cancer at excision. Esserman and colleagues 13 found that in 2 cases of diffuse ALH/LCIS (defined as greater than 1 lobule per core involved by ALH/LCIS) diagnosed at core needle biopsy, invasive lobular carcinoma was found at excision in both cases. Both occurred in cases of LCIS and in both, the invasive lobular carcinomas were not associated with the targeted lesions. On the other hand, of their 24 cases of focal ALH/LCIS (1 involved lobule per core), there were no cases of infiltrating carcinoma and only 1 case of DCIS at a site remote from the sampled LN at excision. 13 They concluded that excision may not be necessary when a diagnosis of only focal LN is made on core biopsy. Similarly, Subhawong et al 27 found that when only minimal ALH, which they defined as 3 or fewer foci, was found at core needle biopsy, excision yielded no cancers. They defined a focus as ALH involvement of 1 to 3 adjacent lobules. Based on these results, they suggested that minimal incidental ALH on core biopsy does not require excision if there is radiologic-pathologic concordance and the means to follow up with imaging. Table 2. Characteristics of Cases in Which There Was an Upgrade to Malignancy at Surgical Excision Core Biopsy Pathology Imaging Finding Excision Pathology LCIS/ALH Architectural distortion IDC on mammography Irregular mass on US LCIS with necrosis Architectural distortion ILC and fibrocystic change on mammography Irregular mass on US Single focus of ALH Spiculated mass ILC and DCIS with fibrocystic change on MRI LCIS with fibrocystic Clumped, segmental ILC change enhancement on MRI Irregular mass on US ALH Irregular mass on MRI ILC Abbreviations: ALH, atypical lobular hyperplasia; DCIS, ductal carcinoma in situ; IDC, invasive ductal carcinoma; ILC, invasive lobular carcinoma; LCIS, lobular carcinoma in situ; MRI, magnetic resonance imaging; US, ultrasound. Arch Pathol Lab Med Quantifying TDLU Holbrook et al 3

5 Table 3. Source, y Calhoun et al, Chaudhary et al, Esserman et al, Karabakhtsian et al, Middleton et al, Rendi et al, Subhawong et al, Previous Publications Analyzing the Extent of Lobular Neoplasia (LN) on Core Biopsy as a Predictor of Upgrade to Malignancy at Surgical Excision No. of Biopsies With Surgical Follow-up Analysis Criteria Results Conclusions 76 LN % involvement by LN on core Greater % involvement correlated with increased chance of upgrade 87 pure ALH, LCIS,.4 total lobules involved 3 upgrades at surgery with or ALH þ LCIS.1 lobule per core no relationship to number of lobules involved 26 LN Diffuse LN:.1 lobule per Diffuse: 2 upgrades to ILC core Focal LN: 1 lobule per Focal: 1/24 upgraded to core DCIS at distant site from biopsy 92 pure ALH, LCIS, or ALH þ LCIS 20 cases of benign LN concordant selected for excision 73 ALH, 33 LCIS (including 25 with ADH) 42 ALH, 12 LCIS (benign, concordant) Total number of TDLUs involved on core biopsy Focal: 2 involved TDLUs Extensive: 3 involved TDLUs,4 TDLUs involved versus 4 TDLUs involved Minimal ALH: 3 foci (focus: ALH in 1 3 adjacent lobules) Extent of involvement may correlate with upgrade on excision No correlation between extent of LN and upgrade rate Excise diffuse LN 10/92 upstaged Number of TDLUs Upstaged: 1 4 TDLUs involved on core biopsy (mean, 2.3 TDLUs) does not predict for upgrade Not upstaged: 1 8 TDLUs (mean, 3.3 TDLUs) No upgrades for focal ALH, LCIS, LN 1/46 upgrades if,4 TDLUs 6/29 upgrades if 4 TDLUs Minimal ALH: 0 upgrades LCIS: 5/12 upgrades Surgical excision for,3 involved TDLUs is not necessary No surgery for calcium lesions in non high-risk patients with,4 TDLUs involved Minimal incidental ALH does not require excision Abbreviations: ADH, atypical ductal hyperplasia; ALH, atypical lobular hyperplasia; DCIS, ductal carcinoma in situ; ILC, invasive lobular carcinoma; LCIS, lobular carcinoma in situ; TDLU, terminal ductal lobular unit. Middleton and colleagues 31 found that if fewer than 3 TDLUs were involved by ALH/LCIS, the risk of upgrade to cancer was less than 2%, and they concluded that surgery was not necessary in these cases. Rendi et al 32 found no cases of upgrade to cancer at excision if fewer than 4 TDLUs were involved. Calhoun et al 33 found that cases averaging 30% involvement by LN on core needle biopsy were upgraded, whereas cases averaging 15% involvement were not. Finally, Aulmann et al 34 followed 88 patients with LCIS on biopsy for 10 years and found that 9 developed subsequent invasive cancer. Of these 9 patients, 4 had extensive LCIS, which they defined as greater than 10 affected lobules in their original biopsy. 34 Other authors, however, have shown no correlation between extent of disease at core biopsy and likelihood of upgrade. For example, Karabakhtsian et al 8 counted TDLUs involved by ALH/LCIS in 92 cases and found that this number had no predictive value for the upgrade rate, with a mean of 2.3 TDLUs involved in upgraded cases, and 3.3 TDLUs involved in cases that were not upgraded. Their upgrade rate ranged from 8% (ALH) to 19% (LCIS). 8 Chaudhary and colleagues 19 also found no correlation between number of TDLUs involved by ALH/LCIS and risk of upgrade in 87 cases. Our results support this latter group in that we also found no correlation between the total number of TDLUs, the number of TDLUs involved by ALH/ LCIS, or the percentage of TLDUs involved by ALH/LCIS in the core biopsy specimen and upgrade to malignancy at surgical excision. In our opinion, quantifying morphologically normal and diseased TDLUs will not aid in triaging patients to surgery or not. However, there may well be specific histologically identifiable morphologic features of LN on core biopsy, including cellular and nuclear pleomorphism and comedonecrosis, that might suggest a higher likelihood of upgrade and the need for surgical excision. Pleomorphic LCIS and LCIS with necrosis are known to be associated with a worse prognosis than classic LCIS. 20,35 In our study, the only case of LCIS with necrosis was associated with an upgrade to malignancy at excision; however, the 4 cases of pleomorphic LCIS were not. Further, continued refinement in analyzing the genomic and immunohistochemical features of lobular neoplasia on core biopsy might also prove helpful in predicting those lesions more likely to presage or accompany frank malignancy. Our study is limited by the retrospective nature of the data collection. Not all patients included in this study had both core biopsy and surgical excision, introducing the possibility of selection bias with regard to which patients received surgery and which did not after a core biopsy diagnosis of ALH/LCIS. Additionally, of the 13 patients who did not have surgery, 5 (38.5%) received some type of chemoprevention. This may have prevented a malignancy from becoming apparent clinically or on imaging in the followup period. Lastly, this study is retrospective in nature. However, this limitation is mitigated somewhat by the blinded and thus prospective radiologic, pathologic, and radiologic-pathologic analysis that was the study design. Also, as in virtually all studies discussing the rare diagnosis of lobular neoplasia, our study involves a relatively small 4 Arch Pathol Lab Med Quantifying TDLU Holbrook et al

6 Figure 1. Radiologic-pathologic discordant mass in a 58-year-old woman undergoing high-risk screening magnetic resonance imaging (MRI). A, Axial T1 postcontrast MRI demonstrates a spiculated, enhancing mass in the left breast (arrow). B, MRI-guided biopsy yielded fibrocystic change with a single focus of atypical lobular hyperplasia. This result was felt to not sufficiently explain the mass, and therefore the case was considered discordant and excision performed (C), yielding invasive lobular carcinoma, Nottingham grade 1, and a focus of ductal carcinoma in situ (arrow) (hematoxylineosin, original magnifications 340 [B] and 3100 [C]). Figure 2. Radiologic-pathologic concordant lesion with no upgrade to malignancy in a 41-year-old woman with a mass noted at screening mammogram. A, Ultrasound of the left breast demonstrates a circumscribed, oval mass (calipers). B, Core needle biopsy reveals convex margin of a fibroadenoma (black arrows) with adjacent atypical lobular hyperplasia (ALH; white arrow). This was felt to be concordant. C, Excision yielded fibroadenoma completely involved by ALH (hematoxylin-eosin, original magnification 340). number of patients. This limits the study power, and thus its generalizability. In conclusion, we found no added value or utility in quantifying normal and abnormal TDLUs to aid in determining whether to recommend surgical excision or not. This study does validate the results of Atkins et al, 21 demonstrating again that assiduous radiologic-pathologic correlation to determine lesion concordance/discordance Arch Pathol Lab Med Quantifying TDLU Holbrook et al 5

7 Figure 3. A 48-year-old woman with extensive lobular carcinoma in situ (LCIS) without upgrade to malignancy. A, Craniocaudal magnification mammogram demonstrates amorphous calcifications (arrow). B, Core biopsy found extensive LCIS. This patient had the greatest total number of abnormal terminal ductal lobular units in our study (38 of 122; 31.1%). C, Despite this, at excision there was no upgrade to malignancy, with only LCIS found (hematoxylin-eosin, original magnification 3100). can predict those patients who would most likely benefit from surgical excision. Certainly, additional studies are required in order for definitive guidelines to be established. References 1. Arpino G, Allred DC, Mohsin SK, Weiss HL, Conrow D, Elledge RM. Lobular neoplasia on core-needle biopsy clinical significance. Cancer. 2004; 101(2): Brem RF, Lechner MC, Jackman RJ, et al. Lobular neoplasia at percutaneous breast biopsy: variables associated with carcinoma at surgical excision. AJR Am J Roentgenol. 2008;190(3): Destounis SV, Murphy PF, Seifert PJ, et al. Management of patients diagnosed with lobular carcinoma in situ at needle core biopsy at a communitybased outpatient facility. AJR Am J Roentgenol. 2012;198(2): Ibrahim N, Bessissow A, Lalonde L, et al. Surgical outcome of biopsyproven lobular neoplasia: is there any difference between lobular carcinoma in situ and atypical lobular hyperplasia? AJR Am J Roentgenol. 2012;198(2): Elsheikh TM, Silverman JF. Follow-up surgical excision is indicated when breast core needle biopsies show atypical lobular hyperplasia or lobular carcinoma in situ: a correlative study of 33 patients with review of the literature. Am J Surg Pathol. 2005;29(4): Polom K, Murawa D, Pawelska A, Murawa P. Atypical lobular hyperplasia and lobular carcinoma in situ without other high-risk lesions diagnosed on vacuum-assisted core needle biopsy: the problem of excisional biopsy. Tumori. 2009;95(1): Shin SJ, Rosen PP. Excisional biopsy should be performed if lobular carcinoma in situ is seen on needle core biopsy. Arch Pathol Lab Med. 2002; 126(6): Karabakhtsian RG, Johnson R, Sumkin J, Dabbs DJ. The clinical significance of lobular neoplasia on breast core biopsy. Am J Surg Pathol. 2007;31(5): Foster MC, Helvie MA, Gregory NE, Rebner M, Nees AV, Paramagul C. Lobular carcinoma in situ or atypical lobular hyperplasia at core-needle biopsy: is excisional biopsy necessary? Radiology. 2004;231(3): Lewis JL, Lee DY, Tartter PI. The significance of lobular carcinoma in situ and atypical lobular hyperplasia of the breast. Ann Surg Oncol. 2012;19(13): Cangiarella J, Guth A, Axelrod D, et al. Is surgical excision necessary for the management of atypical lobular hyperplasia and lobular carcinoma in situ diagnosed on core needle biopsy?: a report of 38 cases and review of the literature. Arch Pathol Lab Med. 2008;132(6): Crisi GM, Mandavilli S, Cronin E, Ricci A Jr. Invasive mammary carcinoma after immediate and short-term follow-up for lobular neoplasia on core biopsy. Am J Surg Pathol. 2003;27(3): Esserman LE, Lamea L, Tanev S, Poppiti R. Should the extent of lobular neoplasia on core biopsy influence the decision for excision? Breast J. 2007; 13(1): Hussain M, Cunnick GH. Management of lobular carcinoma in-situ and atypical lobular hyperplasia of the breast a review. Eur J Surg Oncol. 2011;37(4): Niell B, Specht M, Gerade B, Rafferty E. Is excisional biopsy required after a breast core biopsy yields lobular neoplasia? AJR Am J Roentgenol. 2012;199(4): Nagi CS, O Donnell JE, Tismenetsky M, Bleiweiss IJ, Jaffer SM. Lobular neoplasia on core needle biopsy does not require excision. Cancer. 2008; 112(10): Renshaw AA, Derhagopian RP, Martinez P, Gould EW. Lobular neoplasia in breast core needle biopsy specimens is associated with a low risk of ductal carcinoma in situ or invasive carcinoma on subsequent excision. Am J Clin Pathol. 2006;126(2): Renshaw AA, Cartagena N, Derhagopian RP, Gould EW. Lobular neoplasia in breast core needle biopsy specimens is not associated with an increased risk of ductal carcinoma in situ or invasive carcinoma. Am J Clin Pathol. 2002;117(5): Chaudhary S, Lawrence L, McGinty G, Kostroff K, Bhuiya T. Classic lobular neoplasia on core biopsy: a clinical and radio-pathologic correlation study with follow-up excision biopsy. Mod Pathol. 2013;26(6): Hwang H, Barke LD, Mendelson EB, Susnik B. Atypical lobular hyperplasia and classic lobular carcinoma in situ in core biopsy specimens: routine excision is not necessary. Mod Pathol. 2008;21(10): Atkins KA, Cohen MA, Nicholson B, Rao S. Atypical lobular hyperplasia and lobular carcinoma in situ at core breast biopsy: use of careful radiologicpathologic correlation to recommend excision or observation. Radiology. 2013; 269(2): Bauer VP, Ditkoff BA, Schnabel F, Brenin D, El-Tamer M, Smith S. The management of lobular neoplasia identified on percutaneous core breast biopsy. Breast J. 2003;9(1): Bowman K, Munoz A, Mahvi DM, Breslin TM. Lobular neoplasia diagnosed at core biopsy does not mandate surgical excision. J Surg Res. 2007;142(2): Middleton LP, Grant S, Stephens T, Stelling CB, Sneige N, Sahin AA. Lobular carcinoma in situ diagnosed by core needle biopsy: when should it be excised? Mod Pathol. 2003;16(2): Arch Pathol Lab Med Quantifying TDLU Holbrook et al

8 25. Murray MP, Luedtke C, Liberman L, Nehhozina T, Akram M, Brogi E. Classic lobular carcinoma in situ and atypical lobular hyperplasia at percutaneous breast core biopsy: outcomes of prospective excision. Cancer. 2013;119(5): Liberman L, Sama M, Susnik B, et al. Lobular carcinoma in situ at percutaneous breast biopsy: surgical biopsy findings. AJR Am J Roentgenol. 1999; 173(2): Subhawong AP, Subhawong TK, Khouri N, Tsangaris T, Nassar H. Incidental minimal atypical lobular hyperplasia on core needle biopsy: correlation with findings on follow-up excision. Am J Surg Pathol. 2010;34(6): Berg WA, Mrose HE, Ioffe OB. Atypical lobular hyperplasia or lobular carcinoma in situ at core-needle breast biopsy. Radiology. 2001;218(2): Philpotts LE, Shaheen NA, Jain KS, Carter D, Lee CH. Uncommon high-risk lesions of the breast diagnosed at stereotactic core-needle biopsy: clinical importance. Radiology. 2000;216(3): Ottesen GL, Graversen HP, Blichert-Toft M, Zedeler K, Andersen JA. Lobular carcinoma in situ of the female breast: short-term results of a prospective nationwide study: the Danish Breast Cancer Cooperative Group. Am J Surg Pathol. 1993;17(1): Middleton LP, Sneige N, Coyne R, et al. Most lobular carcinoma in situ and atypical lobular hyperplasia diagnosed on core needle biopsy can be managed clinically with radiologic follow-up in a multidisciplinary setting. Cancer Med. 2014;3(3): Rendi MH, Dintzis SM, Lehman CD, Calhoun KE, Allison KH. Lobular insitu neoplasia on breast core needle biopsy: imaging indication and pathologic extent can identify which patients require excisional biopsy. Ann Surg Oncol. 2012;19(3): Calhoun BC, Collie AM, Lott-Limbach AA, et al. Lobular neoplasia diagnosed on breast core biopsy: frequency of carcinoma on excision and implications for management. Ann Diagn Pathol. 2016;25: Aulmann S, Penzel R, Longerich T, Funke B, Schirmacher P, Sinn HP. Clonality of lobular carcinoma in situ (LCIS) and metachronous invasive breast cancer. Breast Cancer Res Treat. 2008;107(3): Eusebi V, Magalhaes F, Azzopardi JG. Pleomorphic lobular carcinoma of the breast: an aggressive tumor showing apocrine differentiation. Hum Pathol. 1992;23(6): Arch Pathol Lab Med Quantifying TDLU Holbrook et al 7