Significance of flat epithelial atypia on mammotome core needle biopsy: should it be excised? B

Transcription

1 Human Pathology (2007) 38, Original contribution Significance of flat epithelial atypia on mammotome core needle biopsy: should it be excised? B Lakshmi P. Kunju MD, Celina G. Kleer MD* Department of Pathology, University of Michigan School of Medicine, Ann Arbor, MI 48331, USA Comprehensive Cancer Center, University of Michigan School of Medicine, Ann Arbor, MI 48331, USA Received 19 May 2006; revised 27 July 2006; accepted 4 August 2006 Keywords: Flat epithelial atypia; Atypical ductal hyperplasia; Breast core needle biopsy Summary The aim of this study was to determine the morphologic types, associations, and significance of flat epithelial atypia (FEA) with or without atypical ductal hyperplasia (ADH) in mammotome core needle biopsies. We evaluated the correlation of FEA in core biopsies with follow-up excision biopsies to predict the likelihood of upgrade to carcinoma. We also investigated the utility of Ki-67 in predicting which lesions were associated with carcinoma in the excisional biopsies. Core biopsies with a diagnosis of atypia were categorized as pure FEA, pure ADH, or both. The following parameters were recorded: indication for core biopsies, presence of microcalcifications, inflammation, and stromal changes. A total of 60 core biopsies from 56 patients were studied. Pure ADH, pure FEA, and concomitant FEA and ADH were seen in 13%, 23%, and 64% of core biopsies, respectively. The most common architectural pattern of FEA resembled blunt duct adenosis (52%), followed by cystically dilated ducts with secretions (38%) and apocrine features (10%). Chronic inflammation and stromal changes were noted in 29% and 36% of FEA, respectively. Excisional biopsies in 48 of 56 patients demonstrated ductal carcinoma in situ and/or invasive carcinoma in 10 patients (21%), lobular carcinoma in situ or atypical lobular hyperplasia in 5 (11%), residual ADH in 11 (23%), and no atypia in 24 patients (50%). Three (21%) of 14 pure FEA upgraded to ductal carcinoma in situ and/or invasive carcinoma on excisional biopsy. The staining for Ki-67 in FEA/ADH was similar regardless of whether they were upgraded to carcinoma or not. In summary, we found a strong association between FEA and ADH, which may reflect a biologic progression. Most FEAs have a low-power appearance of a well-circumscribed group of ducts. Chronic inflammation and stromal changes are present in a subset of cases. Flat epithelial atypia shows a risk of upgrade to carcinoma similar to that of ADH and, hence, should be recognized and warrants a follow-up excision. D 2007 Elsevier Inc. All rights reserved. B This work was supported in part by National Cancer Institute grants K08CA (CGK) and R01CA (CGK), and Department of Defense grant DAMD (CGK). Presented in part at the 95th United States and Canadian Academy of Pathology (USCAP) Meeting, Atlanta, GA, February * Corresponding author. University of Michigan, Ann Arbor, MI 48109, USA. address: kleer@med.umich.edu (C. G. Kleer) /$ see front matter D 2007 Elsevier Inc. All rights reserved. doi: /j.humpath

2 36 1. Introduction Widespread use of mammography as a screening tool has resulted in increasing numbers of breast biopsies performed for suspicious or indeterminate microcalcifications. On these biopsies, surgical pathologists frequently encounter lesions designated as bflat epithelial atypiaq (FEA), a term introduced by the World Health Organization (WHO) Working Group on the Pathology and Genetics of Tumors of the Breast [1]. This is not a new entity and many authors have previously used a wide variety of names to describe it including bsmall ectatic ducts lined by atypical duct cells with apocrine snoutsq [2], bcolumnar alteration with prominent apical snouts and secretionsq (CAPSSs) [3], batypical cystic lobulesq [4], and bductal intraepithelial neoplasia, flat typeq [5]. Flat epithelial atypia is a descriptive term for an intraductal alteration characterized by replacement of native epithelial cells by a single or three to five layers of monotonous atypical cuboidal to columnar cells with apical snouts. Many spaces contain flocculant or secretory material with microcalcifications. Flat epithelial atypia is distinguished from columnar cell change and columnar cell hyperplasia by the presence of mild cytologic atypia and from atypical ductal hyperplasia (ADH) and ductal carcinoma in situ (DCIS) by the absence of architectural atypia. Whereas a diagnosis of ADH on core needle biopsy specimens is regarded as indication for surgical excision, a diagnosis of FEA on needle biopsy specimens poses dilemmas with regard to the most appropriate clinical management. Recent studies as well as emerging genetic evidence suggest that FEA may be the earliest precursors of low-grade DCIS [1,6]. Assessment of the clinical significance of FEA has been challenging owing to a lack of uniform terminology in the literature as well as the small number of cases that have been studied in a systematic manner. The aim of our study was to determine the histologic types, associations, and significance of FEA in mammotome core needle biopsies. Our goal was also to correlate FEA in core biopsies with the subsequent excision to predict the likelihood of upgrade to DCIS or invasive carcinoma. The utility of the proliferation marker Ki-67 (MIB-1 clone) in predicting which ADH or FEA lesions were associated with carcinoma in excision was also investigated. 2. Materials and methods 2.1. Study population Using a SNOMED search, we identified 60 core biopsies with a diagnosis of atypia performed between January 2000 and December 2005 at our institution from more than 900 core biopsies performed during the same period. Slides were retrieved from the surgical pathology files of the University of Michigan, and all cases were retrospectively evaluated by the authors (LPK and CGK) who were blinded to the follow-up information. Core biopsies with atypical papillary lesions, lobular carcinoma in situ (LCIS), DCIS, and/or invasive carcinoma were excluded from this study. Relevant clinical data including age of patients and biopsy indication were recorded in all cases. The majority of the core biopsies had been performed by dedicated radiologists using the 14- gauge needle and on average had approximately five passes per biopsy Morphologic review L. P. Kunju, C. G. Kleer All core biopsies were categorized as pure FEA (no associated ADH), pure ADH (no associated FEA), or both (concomitant FEA and ADH). Core biopsies were diagnosed as ADH using established criteria according to WHO classification [1] and/or as FEA according to the classification system outlined by Schnitt and Vincent- Salomon [7]. Atypical ductal hyperplasia was characterized by intraductal proliferation of evenly distributed monomorphic cells growing in micropapillae, tufts, fronds, arcades, solid, and cribriform patterns. Flat epithelial atypia lesions were characterized by variably distended terminal duct lobular units in which the native epithelial cells are replaced by one to several layers of a monotonous, atypical cuboidal to columnar cell population with apical snouts, secretory or floccular material in the lumen, often with microcalcifications. Flat epithelial atypia by definition, unlike ADH, has a flat growth pattern with no complex architecture. As part of our routine protocol, eight slides per needle biopsy specimen were cut. First, third, fifth, and seventh slides were used for hematoxylin and eosin (H&E) staining, whereas the intervening four slides were saved for potential immunohistochemistry (IHC). Morphologic parameters evaluated include the following: (a) morphologic patterns of FEA and/or ADH, (b) presence of microcalcifications, (c) presence of chronic inflammation, and (d) presence of stromal changes including fibrosis and myxoid change. Table 1 Characteristics of core biopsies with atypia Pure FEA Pure ADH FEA and ADH No. of biopsies (%) 14 (23%) 8 (13%) 38 (64%) Age (median) (y) Indication for biopsy Microcalcifications 12 (86%) 5 (63%) 29 (76%) Mammographic 2 (14%) 3 (37%) 9 (24%) density Calcifications 12 (86%) 5 (63%) 29 (76%) Inflammation None-mild 10 (71%) 8 (100%) 32 (84%) Moderate-severe 4 (29%) 0 6 (16%) Stromal changes 5 (36%) 0 17 (45%)

3 Significance of flat epithelial atypia on mammotome core needle biopsy 37 Fig. 1 Morphologic patterns of FEA. A and C, FEA with blunt duct adenosis pattern (H&E, original magnifications 200 and 400, respectively). B and D, FEA with cystically dilated glands (H&E, original magnifications 100 and 400, respectively). In both architectural patterns, the cells have a flat growth pattern (no complex architecture, unlike ADH) and show cytologic atypia resembling that seen in lowgrade DCIS. The cells lack polarity with respect to basement membrane and have round to ovoid nuclei with occasional nucleoli Immunohistochemistry Assessment of the proliferation marker Ki-67 (MIB-1 clone) was determined by IHC on all biopsies which showed upgrade to DCIS and/or invasive carcinoma on follow-up excision biopsies as well as on 11 control biopsies which showed no residual atypia on follow-up excision. For this we used a refined labeled streptavidin-biotin technique. After paraffin removal and hydration, slides were treated with Tris (0.25 mol/l)-edta (0.1 mmol/l) buffer, ph 9.0, in a microwave pressure cooker for 15 minutes for optimal antigen retrieval before immunostaining. Sections were stained with a monoclonal antibody against Ki-67 (MIB-1 clone) (Dako Cytomation, Carpinteria, Calif) at 1:50 dilution for 32 minutes at 428C on 4-lm-thick sections obtained from formalin-fixed, paraffin-embedded blocks. A Ventana Basic DAB Detection Kit (Ventana Medical Systems, Tuscon, Ariz) was used according to manufacturer s specifications, and staining was performed on the Ventana Benchmark XT autostainer. Sections were then counterstained with Gills hematoxylin. Both authors (LPK and CGK) interpreted the IHC results and an average score was used for evaluation. Table 2 Follow-up excisional biopsy diagnoses after a core biopsy with ADH and or FEA Core biopsy diagnosis Follow-up excisional biopsy DCIS/invasive Ca ADH LCIS/ALH No atypia No resection Pure FEA (n = 14) 3 (21%) 5 (36%) 1 (8%) 3 (21%) 2 (14%) Pure ADH (n = 8) 3 (38%) 0 2 (25%) 2 (25%) 1 (12%) FEA and ADH (n = 38) 4 (11%) 6 (16%) 2 (5%) 19 (50%) 7 (18%)

4 38 L. P. Kunju, C. G. Kleer 2.4. Follow-up information Follow-up excision biopsies were reviewed for all cases when available. The final diagnosis on excision biopsy was recorded and compared with the biopsy diagnosis. 3. Results 3.1. Needle biopsy results A total of 60 core needle biopsies from 56 patients (age range, years) with a diagnosis of atypia were included in the study. In 3 patients, bilateral biopsies were performed, whereas one patient had 2 biopsies from distinctly separate foci in the same breast. Of these 60 core biopsies, 14 (23%) had pure FEA, 8 (13%) had pure ADH, and the remaining 38 (64%) core biopsies had concomitant FEA and ADH. The majority (46/60, 77%) of core biopsies were performed for microcalcifications detected on screening mammography. The remaining (23%) cases were performed for abnormal mammographic densities. Table 1 summarizes the clinical and pathological characteristics of the patients with atypia in the core biopsy. Of note, when FEA was present, it was frequently seen in association with ADH (64%). A small number of core biopsies contained FEA in isolation (23%). A subset of FEA lesions (with or without ADH) were associated with moderate to severe chronic inflammation composed predominantly of lymphocytes and plasma cells. Furthermore, stromal changes including myxoid areas and fibrosis were seen in association with FEA and ADH in a subset of cases (36% and 45% cases, respectively). The most common morphologic pattern (52%, 27/52 core biopsies) of FEA was that of a lobular well-circumscribed configuration of variably distended terminal ducts (blunt duct adenosis pattern on low power [10]) (Fig. 1A) followed by large cystically dilated ducts with flocculent secretions in lumen (38%, 20/52 core biopsies) (Fig. 1B). The least common morphologic pattern observed was FEA with apocrine features (10%, 5/52 core biopsies). In all cases, regardless of the low-power appearance, the cells lining the acini or the ducts showed low-grade cytologic atypia characterized by epithelial cells with round to ovoid nuclei, slightly increased nuclear-cytoplasm ratio, and disorganization of nuclei relative to basement membrane (Fig. 1C and D). The most common morphologic pattern of ADH was cribriform (57%, 26/46 core biopsies) (Fig. 2A) followed by micropapillary (28%, 13/46) (Fig. 2B). The least common morphologic pattern observed was ADH with apocrine features (13%, 6/46) (Fig. 2C). Follow-up excision biopsies were available in 50 of 60 atypical core biopsies (48/56 patients). Eight patients did not undergo follow-up excision, whereas one patient with bilateral biopsies had a unilateral excision and one patient Fig. 2 Morphologic features of ADH. A, Cribriform ADH (H&E, original magnification 200) characterized by a cribriform proliferation of monomorphic cells with lumens of different sizes and occasional overlapping of cells. B, Micropapillary ADH (H&E, original magnification 200) characterized by a micropapillary proliferation into the lumen of a large duct, not involving the entire circumference of the duct. Note that cells are not as uniform as they would be in low-grade DCIS. FEA is present at the periphery (lower left corner). C, Apocrine ADH (H&E, original magnification 200) characterized by proliferating cells with abundant eosinophilic cytoplasm and cytologic atypia. Note the presence of large nucleoli distinctly visible at low to medium magnification.

5 Significance of flat epithelial atypia on mammotome core needle biopsy 39 with core biopsies from two distinct sites on the same breast underwent a single excision. Overall, of 48 patients who underwent follow-up excision, 10 patients (21%) had DCIS and or invasive carcinoma, 11 (23%) had residual ADH, 5 (11%) had LCIS or atypical lobular hyperplasia (ALH), and 24 patients (50%) had no residual atypia. These results are summarized in Table 2. Of the 10 patients who were upgraded to DCIS or invasive carcinoma on excision biopsy, 3 (21%) had pure FEA on core biopsy. In 8 of these patients (including 3 patients with pure FEA), the indication for core biopsy was microcalcifications, whereas in the remaining two patients the core biopsies were performed for mammographic densities. Of the 3 patients with FEA who developed a more advanced lesion on excision, 2 had invasive carcinoma (one with a Scarff-Bloom-Richardson grade 1 with tubular features, and the other with a Scarff-Bloom-Richardson grade 2). One patient had low nuclear grade DCIS on follow-up excision. Overall, 7 (15%) of 46 patients with ADH with or without FEA demonstrated DCIS/invasive carcinoma on follow-up excision. There was no association between any specific pattern of FEA and or ADH and risk of upgrade to DCIS/invasive carcinoma. Ki-67 expression in FEA and ADH was similar (mean, 3%; range, 0%-10%), regardless of whether they upgraded to DCIS/invasive carcinoma or not on excision. 4. Discussion When a diagnosis of ADH is rendered on core needle biopsy, surgical excision is recommended to avoid the underdiagnosis of cancer [8,9]. The Cancer Committee of the College of American Pathologists has assigned ADH a moderately increased risk (relative risk of 4-5) for subsequent development of invasive breast cancer [1]. After a diagnosis of ADH at biopsy, 3.7% to 22% of women subsequently develop invasive carcinoma [1,10,11]. The rate of underdiagnosis of carcinoma for ADH on core needle biopsy performed with mammotome based on the findings of a subsequent excision ranges from 15% to 39% [12,13]. The identification of FEA, however, especially on core biopsies performed for mammographically detected microcalcifications, poses a diagnostic challenge to surgical pathologists as well as a management dilemma for the clinician. Despite the diagnostic difficulty, studies to assess reproducibility in the evaluation of FEA have demonstrated moderate to substantial interobserver reproducibility after a tutorial [14,15]. In general, close scrutiny at medium to high magnification is recommended to recognize this subtle lowgrade cytologic atypia. Thus, FEA is increasingly recognized and diagnosed by surgical pathologists and the clinical implications of this diagnosis need to be clearly defined. To date, there are little data in the published literature to support either observation or follow-up excision as a definitive line of treatment of FEA. Emerging data suggest that FEA most likely represents the earliest morphologically recognizable precursor of low-grade DCIS [7]. As a result, the diagnosis of FEA might have potential diagnostic as well as therapeutic implications. We present our series of core biopsies with atypia and specifically correlate the associations, types, and significance of FEA in the core biopsy with the subsequent followup excision. All core biopsies had low-grade cytologic atypia and were architecturally bflat,q that is, devoid of all complex architectural patterns including micropapillary tufts, Roman bridges, or rigid arcades which are frequently noted in ADH or DCIS. Several important conclusions can be drawn from the present study. First, although FEA may occur in isolation, in the vast majority of cases FEA is detected in association with ADH, and it may be also associated with DCIS, lobular neoplasia, and invasive carcinoma [3,16-18]. In our series, 64% (38/60) of the core biopsies had concomitant FEA and ADH, whereas only 23% (14/60) of the core biopsies had FEA only without associated ADH. Second, the median age of patients with pure FEA in our series was slightly younger compared to those with ADH with or without FEA (51 versus 53 years and 54 years, respectively). This is similar to a study by Fraser et al [3] in which patients diagnosed with CAPSS with atypia had a median age of 49.5 years. Clinically, patients with pure FEA are asymptomatic and rarely present with a palpable mass. This observation was confirmed in our series in 86% (12/ 14) of cases with pure FEA, the core biopsy was performed because of mammographically detected calcifications versus 76% and 63% patients of ADH with and without FEA, respectively. The presence of microcalcifications was confirmed microscopically in 12 of 14 core biopsies with pure FEA. In only 2 of 14 cases with pure FEA, the core biopsy was performed because of a mammographically detected asymmetry, and in addition to FEA, one of these patients also had ALH in core biopsy. Because of the reported association between chronic inflammation and neoplasia in other organs including the prostate gland and colon [19,20], we set out to determine whether chronic inflammation was associated with FEA and ADH as well. We found that moderate to severe chronic inflammation as well as stromal changes, including myxoid change and fibrosis around the affected ducts, was associated with a subset of cases of ADH and FEA (Table 1). There has been no study to date that has reported an association of inflammation and/or stromal changes with FEA. These findings are intriguing and lead us to postulate that perhaps chronic inflammation and resulting stromal changes may contribute to the development of epithelial atypia, as it has been suggested in other organs [19,20]. This finding warrants further investigation. We found that a large proportion of FEA have a morphologic pattern resembling blunt duct adenosis on low power. Although not as common, cystically dilated

6 40 ducts with secretions in their lumen and apocrine features were other morphologic patterns of FEA and need to be recognized by surgical pathologists. In all patterns, scrutiny at medium to high magnification demonstrates the lowgrade cytologic atypia, resembling cells seen in low-grade DCIS or ADH. The cells lack polarity, that is, are not regularly oriented perpendicular to the basement membrane and usually have round to ovoid nuclei. Among the ADH group, the most common morphologic pattern was that of cribriform ADH followed by micropapillary ADH. The recognition of these low-power patterns is important to make the correct diagnosis. Notably, in our series 3 (21%) of 14 patients with only FEA on the core biopsy had DCIS/invasive cancer on subsequent excision. Despite a small number of cases, our findings are consistent with the very limited available data [18,21,22]. Two of these studies [21,22] have been published in abstract form only and reported a more advanced lesion on subsequent excision in approximately one third of cases. The study by Guerra-Wallace et al [18] found a 13% rate of cancer among 31 patients with CAPSS with atypical features who underwent subsequent excision. However, in their study, core biopsies with CAPSS with atypical features had architectural atypia (including micropapillary tufts, epithelial bridges, and early cribriform formation) in addition to nuclear atypia. Flat epithelial atypia as defined by the WHO Working Group on the Pathology and Genetics of Tumors of the Breast (criteria used to define FEA in our series) lacks the architectural features of ADH or low-grade DCIS such as micropapillary tufting, epithelial bridging, and early cribriform formation. Hence, we believe some of their cases of CAPSS with atypia may also contain concomitant ADH. Our data highlight the importance of recognizing and diagnosing FEA in core needle biopsies. The presence of FEA even in isolation warrants a follow-up excision as it is associated with an increased risk of having carcinoma in the excision biopsy in a group of patients. The discovery of biomarkers that can predict which FEA or ADH lesions are associated with carcinoma in the excision biopsy is highly desirable and may have important clinical implications in the management of these patients. We found that a very low percentage of cells in FEA and/or ADH in our study showed a positive staining for Ki-67, indicating the low proliferation rate of FEA. This finding is consistent with that reported in the literature [23] and highlights that FEA is a slowly proliferating lesion. The low Ki-67 also suggests that this marker is not useful in the diagnosis of FEA or in the prediction of concomitant or subsequent carcinoma. In summary, there is a strong association between FEA and ADH, which may reflect a biologic progression. Most FEAs have a morphologic appearance resembling blunt duct adenosis on low power and are often associated with stromal changes and chronic inflammation. Based on our study, FEA shows a risk of upgrade to carcinoma similar to that of ADH. Thus, FEA needs to be recognized and warrants a follow-up excision. Larger as well as comprehensive outcome studies with long follow-up periods are required to characterize the natural history of FEA and better define patient management. Acknowledgments The authors would like to thank Robin Kunkel for imaging assistance. References L. P. Kunju, C. G. Kleer [1] Tavassoli F, Schnitt SJ, Hoefler H, et al. Intraductal proliferative lesions. In: Tavassoli FA, Devilee P, editors. World Health Organization classification of tumors of the breast and female genital organs. Lyon (France)7 IARC Press; p [2] Goldstein NS, O Malley BA. Cancerization of small ectatic ducts of the breast by ductal carcinoma in situ cells with apocrine snouts: a lesion associated with tubular carcinoma. Am J Clin Pathol 1997;107: [3] Fraser JL, Raza S, Chorny K, Connolly JL, Schnitt SJ. Columnar alteration with prominent apical snouts and secretions: a spectrum of changes frequently present in breast biopsies performed for microcalcifications. Am J Surg Pathol 1998;22: [4] Oyama T, Maluf H, Koerner F. Atypical cystic lobules: an early stage in the formation of low-grade ductal carcinoma in situ. Virchows Arch 1999;435: [5] Tavassoli FA. Ductal carcinoma in situ: introduction of the concept of ductal intraepithelial neoplasia. Mod Pathol 1998;11: [6] Simpson PT, Gale T, Reis-Filho JS, et al. Columnar cell lesions of the breast: the missing link in breast cancer progression? A morphological and molecular analysis. Am J Surg Pathol 2005;29: [7] Schnitt SJ, Vincent-Salomon A. Columnar cell lesions of the breast. Adv Anat Pathol 2003;10: [8] Winchester DJ, Bernstein JR, Jeske JM, et al. Upstaging of atypical ductal hyperplasia after vacuum-assisted 11-gauge stereotactic core needle biopsy. Arch Surg 2003;138: [discussion 622-3]. [9] Rao A, Parker S, Ratzer E, Stephens J, Fenoglio M. Atypical ductal hyperplasia of the breast diagnosed by 11-gauge directional vacuumassisted biopsy. Am J Surg 2002;184:534-7 [discussion 537]. [10] 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: [11] Krieger N, Hiatt RA. Risk of breast cancer after benign breast diseases. Variation by histologic type, degree of atypia, age at biopsy, and length of follow-up. Am J Epidemiol 1992;135: [12] 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: [13] Reynolds HE, Poon CM, Goulet RJ, Lazaridis CL. Biopsy of breast microcalcifications using an 11-gauge directional vacuum-assisted device. AJR Am J Roentgenol 1998;171: [14] O Malley FP, Mohsin SK, Badve S, et al. Interobserver reproducibility in the diagnosis of flat epithelial atypia of the breast. Mod Pathol 2006;19: [15] Tan PH, Ho BC, Selvarajan S, Yap WM, Hanby A. Pathological diagnosis of columnar cell lesions of the breast: are there issues of reproducibility? J Clin Pathol 2005;58: [16] Schnitt SJ. The diagnosis and management of pre-invasive breast disease: flat epithelial atypia classification, pathologic features and clinical significance. Breast Cancer Res 2003;5:263-8.

7 Significance of flat epithelial atypia on mammotome core needle biopsy 41 [17] Bratthauer GL, Tavassoli FA. Assessment of lesions coexisting with various grades of ductal intraepithelial neoplasia of the breast. Virchows Arch 2004;444: [18] Guerra-Wallace MM, Christensen WN, White Jr RL. A retrospective study of columnar alteration with prominent apical snouts and secretions and the association with cancer. Am J Surg 2004;188: [19] Sugar LM. Inflammation and prostate cancer. Can J Urol 2006; 13(Suppl 1):46-7. [20] Ying L, Marino J, Hussain SP, et al. Chronic inflammation promotes retinoblastoma protein hyperphosphorylation and E2F1 activation. Cancer Res 2005;65: [21] Harigopal M, Yao DX, Hoda SA, DeLellis RA, Vazquez MF. Columnar cell alteration diagnosed on mammotome core biopsy for indeterminate microcalcifications: results of subsequent mammograms and surgical excision [abstract]. Mod Pathol 2002;15:36A. [22] Brogi E, Tan LK. Findings at excisional biopsy performed after identification of columnar cell change of ductal epithelium in breast core biopsy [abstract]. Mod Pathol 2002;15:29A- 30A. [23] Fraser JL, Raza S, Chorny K, Connolly JL, Schnitt SJ. Immunophenotype of columnar alteration with prominent apical snouts and secretions (CAPSS) [abstract]. Lab Invest 2000;80:21A.