Columnar Cell Lesions of the Breast and Their Differential Diagnosis Anne Vincent-Salomon, MD Department of Pathology, Institut Curie, Paris, France

Case History A 51 year old woman, with no previous breast history, presented a new tightly clustered group of irregular microcalcifications, located in the upper and external part of her left breast, on a systematic mammography. The microcalcifications were classified as Bi-rad 4. The breasts were clinically normal. An image-guided macrobiopsy was decided and performed. Figure 1 Click to view with ImageScope Click to view with a Web-Based Viewer Diagnosis : Atypical hyperplasia in a context of flat epithelial atypia or columnar cell metaplasia with cytologic atypia. Excision is required. Outcome: On the surgical specimen a diagnosis of low grade DCIS in a context of flat epithelial atypia was made. Margins were clear. The patient underwent radiation therapy. Follow-up at 3 years shows no evidence of recurrence. The recent OMS 2003 classification of in situ epithelial lesions of the breast identified a group of lesions named "flat epithelial atypia" (FEA). FEA encompasses columnar cell lesions with atypia without hyperplasia and columnar cell lesions with atypia and with hyperplasia. A surgical resection of the residual microcalcifications for patients with FEA observed on macrobiopsies is recommended in the literature to ovoid under diagnosis of a more pejorative lesion [1, 2, 3] . But FEA encompasses a wide spectrum of lesions and the definition of atypia is still matter of debate and discrepancies among pathologists. Those discrepancies can therefore induce huge differences in patient's management from one center to another. In order to decrease those discrepancies, the use of more precisely defined criterias is warranted. Indeed a recent study by O'Malley et al [4] illustrated the improvement of reproducibility when using standardised diagnosis criteria to define FEA. In addition, in order to progress in the identification of the molecular alterations of FEA, reproducible criteria should be defined and easily used, to guarantee the reproducibility of the results of biological studies such as the very interesting data published by Simpson et al [5] . Histology: Columnar lesions with atypia but without hyperplasia: This group of lesion is described in the OMS classification published in 2003 as ''the replacement of the native epithelial cells by a single layer of midly atypical cells often with apical snouts….". The characteristics of the "midly atypical cells" are not precisely defined. The GEFPICS* group worked on a series of cases of FEA in order to reach a consensus on criteria used to define cytologic atypia in that context (manuscript under preparation). The following criteria can be present in a single lesion but the presence of all criteria is not warranted to assigned ''atypical'' to the observed lesions: One cell thickness is present around the ducts (there is no hyperplasia). The cells are columnar or cuboïdal and taller with a larger cytoplasm than normal cells. The cells have a monotonous aspect. The apical cytoplasmic border forms a circle line, giving an aspect of a second inner circle within the duct. This aspect can even be seen if apical snouts are present. The nuclei are polarised and located at the basal part of the cell if the cell is columnar or at the central part of the cell if the cell is cuboïdal. The nuclei are enlarged and the nuclei / cytoplasm ratio is increased. The epithelial nuclei are taller than the myoepithelial nuclei. The nuclei are round, with a visible nucleoli and an irregular chromatin. The nuclei are regularly spaced, probably because the cytoplasms are larger than those of normal cells. Columnar lesions with atypia and with hyperplasia In the OMS classification these lesions are defined such as "proliferation of a monotonous atypical cells population in the form of stratification of uniform, cuboidal to columnar cells generally up to 3 to 5 cell layers with occasional mounding". The GEFPICS consensus criteria are the following: 2 to 5 cells layers surround the ducts. The cells are monotonous, are larger and taller than normal cells. These cells harbour an elongated and ovoid nucleus and the nuclei are larger than the myoepithelial cell's nuclei. The nuclei are arranged preferentially at the basal part of the cells. There is no nucleolus and the chromatin remains generally thin. The cells are arranged in tufts or in papillae with a large base. At the bottom part of the tufts, the nuclei are superposed like gathered fingers of one hand Some tufts can be floating within the lumens. In the lumen, floculus secretions are often present sometimes being difficult to distinguish from real necrosis. Our group named this secretion ''secrose'' to illustrate that ambiguous aspect. In the FEA spectrum, there is no real necrosis; there is not architectural complex pattern of the hyperplastic cells such as bridges, micropapillae with a narrow base, cribriform arrangements of the proliferative cells around gland lumen. If those aspects are observed, the lesion will be classified as an atypical ductal hyperplasia or ductal carcinoma in situ. It is important to note that it is not uncommon for various combinations of columnar cell lesions to co-exist in the same breast, in the same terminal ducal lobular unit, and even within the same space. Therefore, these diagnoses should not be considered mutually exclusive. In particular, while the lesions we have designated columnar cell change with atypia and columnar hyperplasia with atypia may be observed in isolation, they often co-exist with areas that fulfill the diagnostic criteria for ADH or DCIS, most often of low nuclear grade but sometimes of intermediate nuclear grade; in fact, their presence should prompt a diligent search for such areas. Immunophenotype The cells of the FEA lesions are: Cytokeratin 8/18 positive High molecular weight cytokeratins (cytokeratins 5 and 6 negative) (Mac Grogan et al manuscript in preparation). Estrogen receptor positive [6, 7]. Bcl-2 positive [6]. Differential Diagnosis 1. Apocrine lesions: columnar cell change and columnar cell hyperplasia must be distinguished from benign apocrine lesions such as apocrine metaplasia and apocrine hyperplasia. Morphology : Although the cells comprising both columnar cell lesions and apocrine lesions feature apical cytoplasmic snouts, the cells of apocrine lesions possess more abundant, granular, eosinophilic cytoplasm. In addition, the nuclei of columnar cell lesions at this end of the spectrum tend to be ovoid, whereas those of apocrine lesions tend to be round and often have a single prominent nucleolus. Furthermore, whereas hobnail type cells, highly exaggerated apical snouting, and flocculent intraluminal secretions are seen in some columnar cell lesions, they are not seen in apocrine lesions. Immunophenotype: In contrast to columnar cells, apocrine epithelial cells characteristically lack expression of ER and Bcl2 [6]. 2. Usual cysts: some cases of columnar cell change with atypia may be overlooked entirely on low power microscopic examination due to the lack of significant cellular proliferation and the subtle nature of the cytologic atypia [8]. In fact, TDLUs exhibiting this alteration are often misinterpreted on low power examination as either normal or as showing only microcysts, and it is only after examination of such foci under high magnification that the subtle cytologic atypia becomes evident. 3. Usual Hyperplasia can be developed on columnar cell changes. To ovoid overdiagnosis of ADH, some stainings with cytokeratins 5 and 6 are very helpful as the UDH are stained by CK5/6 [5]. 4. The differential diagnosis for columnar cell lesions with the combination of cellular proliferation and cytologic atypia includes ADH and DCIS. As noted earlier, it is most prudent to categorize columnar cell lesions that show both cytologic atypia and complex architectural patterns as either ADH or DCIS, depending upon the severity and extent of the cytologic and architectural features. For lesions in which the histological features fall short of these criteria, the term columnar cell hyperplasia with atypia or flat epithelial atypia can be used. Reproducibility Clinical Significance A number of studies have provided evidence for a relationship between some columnar cell lesions (i.e., those with atypia) and low-grade DCIS as well as between some columnar cell lesions and invasive breast cancer, particularly tubular carcinoma. This evidence includes the coexistence of these lesions in the same breast, as well as cytologic, immunophenotypic, and genetic similarities between these lesions. Coexistence in same breast In a study of columnar cell lesions termed "columnar alteration with prominent apical snouts and secretions" (CAPSS), Fraser et al noted in some cases a spectrum of changes ranging from CAPSS without cytologic or architectural atypia, to CAPSS with cytologic and/or architectural atypia, to diagnostic areas of low grade DCIS. In addition, among 16 cases in that study in which both CAPSS and DCIS co-existed, the two lesions were located in either the same or adjacent terminal duct lobular units in 81% of the cases and the DCIS lesions in these cases were most often of low (56%) or intermediate (25%) nuclear grade, further suggesting a relationship between these lesions [6] Weidner described a lesion composed of small ectatic ducts lined by one or two layers of columnar cells with apical snouts similar to those seen in tubular carcinoma [9]. He considered this columnar cell lesion to represent an example of low-grade ductal carcinoma in situ. In a more detailed analysis, Goldstein and O'Malley found an association between an atypical columnar cell lesion they called "small ectactic ducts lined by atypical cells with apocrine snouts" with both low-grade DCIS and tubular carcinoma [10].These authors studied 32 tubular carcinomas, 41 grade 1 invasive ductal carcinomas with DCIS, 40 grade 3 invasive ductal carcinomas, 20 low-grade DCIS, and 80 cases with fibrocystic changes. They found the atypical columnar cell lesion in 17 cases, 14 of which were tubular carcinomas and three of which were grade 1 invasive ductal carcinomas. Other authors have also noted an association between various columnar cell lesions and DCIS and/or invasive carcinoma. Page described a spectrum of cytological and architectural changes in a columnar cell lesion they termed "hypersecretory hyperplasia with atypia" At the low end of this spectrum are lesions that have features that overlap with those of columnar alteration of lobules. At the other end of the spectrum are lesions that have architectural and cytologic features that qualify for the diagnosis of ADH or DCIS. In between these two ends of the spectrum are lesions that have some cytologic and/or architectural atypia but the features are not sufficiently well developed to fulfill the diagnostic criteria of either ADH or DCIS. Rosen observed that patients with tubular carcinoma often have foci of lesions he has termed columnar cell or pretubular hyperplasia distributed in the surrounding breast tissue or merging with the carcinoma. This suggested to him that the tubular carcinoma "might sometimes arise when the hyperplastic lesion transformed into intraductal carcinoma." A number of authors have noted an association between some columnar cell lesions and lobular neoplasia (lobular carcinoma in situ and atypical lobular hyperplasia) [11]. Cytologic, immunophenotypic, Some investigators have noted that the cells comprising some columnar cell lesions are cytologically similar or identical to the cells comprising some forms of DCIS or to the cells comprising the glands of tubular carcinoma [8, 10]. Based on the cytologic and immunophenotypic similarity between atypical cystic lobules and low-grade DCIS as well as on their geographic proximity, these authors concluded that atypical cystic lobules represent an early phase in the development of some forms of low-grade DCIS. Tremblay et al [12]showed that the ER were heterogeneously expressed in columnar cell lesions meanwhile, their expression became homogeneous in columnar cell hyperplasia. Genetic similarities In a genetic study of 13 cases of the atypical columnar cell lesion designated "DIN-flat monomorphic type,» Moinfar et al found that 70% showed LOH at one or more of the eight loci evaluated and further, that the genetic alterations in these columnar cell lesions were the same as those in the associated DCIS or invasive cancer [8].However, these authors did not evaluate genetic changes in columnar cell lesions without associated DCIS or invasive carcinoma. Simpson et al [5] in 2003 performed a IHC and a CGH analysis of columnar cell lesions classified according to the classification proposed by Schnitt and Vincent-Salomon in 2003 [1]. Their conclusions were that in columnar cell lesions, chromosomal alterations were clonal and that the mean number alterations detected by CGH increased from the columnar cell lesions without atypia to the columnar cell hyperplasia with atypia. In addition, they observed that CCL shared some of the chromosome alterations observed in ADH or DCIS well differentiated. Dabbs et al in Modern Pathology in 2006 [13], studied loss of heterozygoty in three cases that presented the complete spectrum of lesions from columnar cell changes, columnar cell hyperplasia, to ADH, DCIS and invasive carcinoma. They showed that LOH were rare in columnar cell lesion, were observed in columnar cell lesion with atypia and were more frequently present in ADH, DCIS or invasive carcinomas. Outcome studies To date, only two follow-up studies have directly addressed the clinical significance of columnar cell lesions with atypia. In a review of over 9000 breast biopsies that were initially considered benign, Eusebi et al retrospectively identified 25 patients with so-called "clinging carcinoma" of the flat, monomorphic (low nuclear grade) type [14]. Only one of these patients (4%) is reported to have developed a "local recurrence" after an average follow-up period of 19.2 years. However, the "local recurrence" in this patient consisted of a "clinging carcinoma" lesion histologically identical to the original lesion, and it is not possible to determine whether this simply reflected persistence of the original lesion due to inadequate excision or a true local recurrence. Of note, none of these 25 patients developed an invasive breast cancer within the follow-up period. In another study, 59 patients with "clinging carcinoma" of the low nuclear grade type were identified among the patients entered into European Organization for Research and Treatment of Cancer (EORTC) 10853, a randomized clinical trial comparing excision and radiation therapy and excision alone for the treatment of women with DCIS [15]. There have been no local recurrences among those 59 patients with a median follow-up of 5.4 years. Thus, the very limited available data suggest that among patients with so-called "clinging carcinoma" of the low nuclear grade/monomorphic type (lesions which we would categorize as columnar cell change with atypia or columnar cell hyperplasia with atypia), the likelihood of progression to invasive breast cancer is exceedingly low, at least with the available follow-up. In a study addressing this question in a somewhat different manner, Shaaban et al performed a case-control study of almost 700 patients to determine the relationship between various benign breast lesions and the risk of subsequent breast cancer [16]. The median follow-up period was 5.5 years. In that study, columnar cell lesions (which they referred to as blunt duct adenosis, BDA) were divided into six sub-groups (BDA, NOS; BDA with calcifications; BDA with columnar cell metaplasia; BDA with atypical columnar cell metaplasia; BDA with hyperplasia of the usual type; and BDA with ADH). While the finding of blunt duct adenosis, NOS was associated with about a 2-fold increase in breast cancer risk, no other subgroup of columnar cell lesions was significantly associated with the subsequent development of breast cancer (including those with atypia). However, given that the number of patients in some of these subgroups was quite small and the follow-up period in this study was relatively short, the clinical significance of these findings is not clear. Thus, it should be apparent that additional clinical follow-up studies are needed to better understand the relationship between the various columnar cells lesions and the risk of subsequent breast cancer. Practical Considerations The appropriate pathology work-up and clinical management of patients whose biopsy specimens show columnar cell lesions are evolving as information regarding these lesions accumulates. Recommendations for the work-up and management of these lesions when encountered in either core needle biopsy (CNB) or excision specimens are presented in Table 1. Core needle biopsy The limited available data suggest that neither additional pathology work-up nor excision are required when either columnar cell change or columnar cell hyperplasia without atypia are encountered in a CNB specimen. In contrast, recent data have suggested that when a columnar cell lesion with atypia is encountered in a CNB specimen, subsequent excision shows a more advanced lesion in about one-third of cases, which is sufficiently frequent to recommend excision as a matter of routine in such cases [1]. Excisional biopsy There are no data to suggest that the presence of columnar cell change or columnar cell hyperplasia without atypia in an excisional biopsy specimen requires further pathologic evaluation or additional treatment. However, the presence of a columnar cell lesion with atypia in an excisional biopsy specimen should prompt a careful search for areas with diagnostic features of ADH or DCIS by obtaining additional levels from the block or blocks containing the lesion and by the submission of the remainder of the tissue for histological examination. There are several other considerations regarding atypical columnar cell lesions in excisional biopsy specimens that merit discussion. When a proliferation that fulfills the diagnostic criteria for ADH or DCIS is found to arise in a background of a columnar cell lesion with atypia, it seems most prudent to manage the patient as one would manage ADH or DCIS in any other setting. However, there are two issues that remain to be resolved when a columnar cell lesion with atypia is found to co-exist with diagnostic areas of DCIS, particularly in cases in which the cytologic features of the cells comprising the atypical columnar cell lesion are similar or identical to those of the cells comprising the diagnostic areas of DCIS. The first is whether or not the atypical columnar cell lesion should be taken into consideration in determining the size or the extent of the DCIS lesion, and the second is whether or not the presence of the atypical columnar cell lesion at the excision margins is sufficient to consider the margins "positive," requiring further surgical resection. Given that the available clinical data, albeit limited, suggest that lesions we would classify as columnar cell change with atypia and columnar cell hyperplasia with atypia are associated with a very low risk of recurrence or progression to invasive carcinoma, these lesions should not be taken into consideration when determining the size of a co-existent DCIS lesion or in the evaluation of the status of the margins of excision, even when they are composed of cells that are cytologically similar to those in the diagnostic areas of DCIS. [1, 17] Conclusion : Based on the foregoing observations, it is reasonable to conclude that columnar cell lesions, particularly those with atypia, are likely neoplastic proliferations that may well represent either a precursor of, or the earliest morphologic manifestation of, low/intermediate grade DCIS as well as a precursor to invasive carcinoma, particularly tubular carcinoma. On a daily practice, the use of precise criteria to define FEA will provide, when they are seen on macrobiopsies, the best conditions to evaluate the risk of more pejorative lesions that could be found on consecutive surgical resections and when they are observed on surgical specimens, to determine the level of risk to develop, on a long term follow-up basis, a non high grade DCIS Table 1. Work-up and management recommendations for columnar cell lesions Type of Columnar Cell Lesion Work-up and Management When Found in Core Needle Biopsy Specimen Work-up and Management When Found in Excisional Biopsy Specimen Columnar cell change Columnar cell hyperplasia No additional levels Excision not required No additional levels No further treatment required Columnar change with atypia Columnar cell hyperplasia with atypia Excision required Obtain multiple levels to look for features diagnostic of ADH or DCIS Submit all tissue ? Margin evaluation * GEFPICS: Groupe d'Etude des Facteurs Pronostiques par Immunohistochimie dans les Cancers du Sein , a group of pathologists from the Fédération Française des Centres de Lutte Contre le Cancer ( FFLCNCC). References Schnitt SJ, Vincent-Salomon A. Columnar cell lesions of the breast. Adv Anat Pathol 2003;10(3):113-24. Guerra-Wallace MM, Christensen WN, White RL, Jr. A retrospective study of columnar alteration with prominent apical snouts and secretions and the association with cancer. Am J Surg 2004;188(4):395-8. 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(9):1095-110. 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(2):172-9. Simpson P, Gale T, Reis-Filho J, et al. Columnar Cell Lesions of the Breast: The Missing Link in Breast Cancer Progression?: A Morphological and Molecular Analysis. American Journal of Surgical Pathology 2005;29:734-46. 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(12):1521-7. Allred DC, O'Connell P, Fuqua SA. Biomarkers in early breast neoplasia. J Cell Biochem Suppl 1993;17G:125-31. Moinfar F, Man YG, Bratthauer GL, Ratschek M, Tavassoli FA. Genetic abnormalities in mammary ductal intraepithelial neoplasia-flat type ("clinging ductal carcinoma in situ"): a simulator of normal mammary epithelium. Cancer 2000;88(9):2072-81. Weidner N. Malignant breast lesions that may mimic benign tumors. Semin Diagn Pathol 1995;12(1):2-13. 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(5):561-6. 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):291-9. Tremblay G, Deschenes J, Alpert L, Quenneville LAM. Overexpression of Estrogen Receptors in Columnar Cell Change and in Unfolding Breast Lobules. The Breast Journal 2005;11(5):326-32. Dabbs DJ, Carter G, Fudge M, Peng Y, Swalsky P, Finkelstein S. Molecular alterations in columnar cell lesions of the breast. Mod Pathol 2006;19(3):344-9. Eusebi V, Feudale E, Foschini MP, et al. Long-term follow-up of in situ carcinoma of the breast. Semin Diagn Pathol 1994;11(3):223-35. Bijker N, Peterse JL, Duchateau L, et al. Risk factors for recurrence and metastasis after breast-conserving therapy for ductal carcinoma-in-situ: analysis of European Organization for Research and Treatment of Cancer Trial 10853. J Clin Oncol 2001;19(8):2263-71. Shaaban AM, Sloane JP, West CR, et al. Histopathologic types of benign breast lesions and the risk of breast cancer: case-control study. Am J Surg Pathol 2002;26(4):421-30. 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(5):263-8.