Skip to main content
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2022 Dec 8;82(1):140–161. doi: 10.1111/his.14799

Pitfalls in breast pathology

Cecily Quinn 1,2,, Aoife Maguire 1, Emad Rakha 3
PMCID: PMC10107929  PMID: 36482276

Abstract

Accurate pathological diagnosis is the cornerstone of optimal clinical management for patients with breast disease. As non‐operative diagnosis has now become the standard of care, histopathologists encounter the daily challenge of making definitive diagnoses on limited breast core needle biopsy (CNB) material. CNB samples are carefully evaluated using microscopic examination of haematoxylin and eosin (H&E)‐stained slides and supportive immunohistochemistry (IHC), providing the necessary information to inform the next steps in the patient care pathway. Some entities may be difficult to distinguish on small tissue samples, and if there is uncertainty a diagnostic excision biopsy should be recommended. This review discusses (1) benign breast lesions that may mimic malignancy, (2) malignant conditions that may be misinterpreted as benign, (3) malignant conditions that may be incorrectly diagnosed as primary breast carcinoma, and (4) some IHC pitfalls. The aim of the review is to raise awareness of potential pitfalls in the interpretation of breast lesions that may lead to underdiagnosis, overdiagnosis, or incorrect classification of malignancy with potential adverse outcomes for individual patients.

Keywords: breast cancer mimics, immunohistochemistry pitfalls, overdiagnosis, underdiagnosis

Introduction

‘Overdiagnosis is more common than underdiagnosis, especially by pathologists with less than 10 years’ experience’ and ‘never use a microscope without a low power objective’, written by John Azzopardi in his textbook “Problems in Breast Pathology”, 1 published in 1979. Over 40 years later, in this era of personalized therapy, molecular pathology and multigene testing, digital microscopy, and artificial intelligence, the importance of correct diagnosis at the outset of a patient's journey has never been more important. Non‐operative diagnosis has become the standard of care, such that the pathologist makes far‐reaching decisions on small tissue samples. Careful appraisal of haematoxylin and eosin (H&E)‐stained sections, combining pattern recognition and analytical thinking, supported by the appropriate use of immunohistochemistry (IHC) and other tests, continues to be the mainstay of histological diagnosis. This review aims to highlight pitfalls in the interpretation of breast lesions that may variously lead to underdiagnosis, overdiagnosis, or incorrect classification of malignancy, with potential adverse outcomes for individual patients. The review focusses on interpretation of H&E and IHC in breast pathology and aims to complement existing literature on the subject. 2 , 3 , 4 , 5 Molecular pathology characteristics of individual entities are not discussed. Similarly, discussion of pitfalls related to non‐operative diagnostic procedures, suboptimal radiology‐pathology correlation, and the quality of sampling at gross examination are beyond the scope of this text.

Entities that may result in overdiagnosis of malignancy

Some benign breast lesions have an ‘infiltrative’ appearance that mimics invasion and may be misinterpreted as invasive carcinoma (IC). These include nipple adenoma, benign syringomatous tumour of the nipple, radial scar / complex sclerosing lesion, sclerosing adenosis, and microglandular adenosis. Benign intraduct papilloma may show florid usual ductal hyperplasia (UDH), reactive epithelial changes or sclerosis that may be mistaken for a malignant tumour.

Other lesions at risk for overdiagnosis of malignancy in breast pathology, particularly on core needle biopsy (CNB) specimens, include those that lack the typical epithelial‐myoepithelial glandular bi‐layer, e.g. pleomorphic adenoma and sclerosed adenomyoepithelioma that may be misdiagnosed as metaplastic breast carcinoma (MBC), 6 the epithelioid variant of myofibroblastoma, composed of dyscohesive oestrogen receptor (ER)‐positive epithelioid cells, which resembles invasive lobular carcinoma (ILC), 7 fat necrosis with a florid histiocytic reaction that may be mistaken for apocrine ILC, 8 and Toker cell hyperplasia of the nipple that may be misdiagnosed as Paget's disease of the nipple. Here we describe some entities that may result in overdiagnosis of malignancy based on the histological findings.

Glandular, sclerosing, and papillary lesions

Nipple adenoma

Nipple adenoma, also known as erosive papillomatosis of the nipple, develops in the superficial aspect of the nipple and may present as nipple erosion or ulceration, resembling Paget's disease clinically, particularly if long‐standing. It may occur in females and males and is most common in women in their fifth decade. Focussed clinical examination usually demonstrates a dermal nodule and the astute clinician will recognize the entity. Ultrasound may demonstrate a mass but is often non‐specific.

Histologically, nipple adenoma may be mistaken for IC due to the irregular outline of the glandular proliferation, often involved by florid UDH with foci of necrosis, mimicking ductal carcinoma in situ (DCIS) (Figure 1 ). Malignancy arising in a nipple adenoma has been reported but is uncommon. 9 , 10 Awareness of the condition and the typical clinical presentation usually leads to the correct diagnosis. In doubtful cases, myoepithelial cell (MEC) IHC can be used to highlight the benign nature of the glands and to characterise an accompanying epithelial cell proliferation. In some lesions, the MEC layer may be attenuated or appear to be absent focally and it is advisable to use at least two MEC markers in difficult cases (see section on Myoepithelial cell immunohistochemistry).

Figure 1.

Figure 1

(AD) Nipple adenoma with florid UDH (H&E).

Syringomatous tumour of the nipple

Syringomatous tumour of the nipple, which occurs in the same location as nipple adenoma, is considered an indolent, locally infiltrative, lesion that is typically small in size with potential for local recurrence but does not metastasise. It shows some morphological overlap with low‐grade adenosquamous carcinoma. 11 These lesions may be difficult to distinguish on CNB and knowledge of the precise anatomical location may be helpful. Syringomatous tumour is typically located superficially and may infiltrate the smooth muscle of the areola but rarely extends into the dermis. 12 The tumour is composed of small nests and cords with a variable number of squamous/squamoid cells that are centrally located and may form small keratin‐filled cysts. Individual cells are bland and mitoses are infrequent. In contrast, low‐grade adenosquamous carcinoma arises deeper in the interstitial tissue of the breast, may reach a large size, rarely infiltrates the nipple, and may also feature a neoplastic spindle cell component. Immunophenotypically, the predominant cell type of syringomatous tumour is myoepithelial with expression of p63, high molecular weight (HMW) cytokeratins (CKs), 13 and other MEC markers.

Radial scar/complex sclerosing lesion

Radial scar/complex sclerosing lesions are rarely symptomatic and are most commonly detected at screening mammography. 14 Due to their stellate configuration, these lesions may resemble IC on imaging, although the presence of a radiolucent centre is reported to strongly favour radial scar. The majority of lesions are categorised as level 4 (suspicious for malignancy) using the Breast Imaging Reporting and Data System (BIRADS). 15

Macroscopically, these lesions may not be visible or may produce a spiculate appearance, again mimicking IC. Microscopically, the proliferation of tubular/glandular structures, with the typical radial arrangement, may create an impression of invasive tubular carcinoma, particularly at low power or on CNB. Appreciation of the central fibro‐elastotic nidus, the lack of extension of tubules into peripheral fat, the absence of cytological atypia, and the presence of two cell types should lead to the correct diagnosis. Due to the collapsed nature of the tubules, it may be difficult to identify the peripheral MEC layer on H&E. IHC is usually helpful (Figure 2 ) but MECs may also be difficult to visualise due to cellular attenuation, especially on p63 nuclear stain (see section on Myoepithelial cell immunohistochemistry).

Figure 2.

Figure 2

(A) Radial scar in which entrapped and collapsed tubules are invested by an attenuated layer of MECs (H&E). (B) MECs may be difficult to appreciate even on IHC (p63 stain).

Accompanying benign change, including epithelial hyperplasia, cyst formation, papillomatosis, and sclerosing adenosis, is common. A more significant lesion, e.g. flat epithelial atypia, atypical ductal hyperplasia, DCIS, or invasive breast carcinoma (IBC) occurs in up to one‐third of cases. Radial scar/complex sclerosing lesion, diagnosed on CNB, should be evaluated in accordance with standard criteria. 16

The proliferative phase of radial scar/complex sclerosing lesions may show adenosquamous proliferation that, when florid, may raise the suspicion of low‐grade adenosquamous carcinoma. 17 In such cases, surgical excision is recommended to allow examination of the surrounding breast tissue, which is typically involved by the infiltrating adenosquamous nests in low‐grade adenosquamous carcinoma.

Sclerosing adenosis

Sclerosing adenosis is a benign lobulo‐centric lesion condition that is frequently seen as an incidental finding in breast biopsies and excisions performed for other breast conditions and as a component of radial scar/complex sclerosing lesion. Multiple lobules involved by sclerosing adenosis may coalesce to form a mass lesion, sometimes referred to as ‘nodular adenosis’.

Affected lobules are usually enlarged, with increased numbers of glandular structures that are compressed, elongated, and distorted due to the accumulation of stromal collagen. Individual glands are invested by epithelial cells and MECs, with frequent hyperplasia of the latter. Early lesions tend to be cellular and may produce a spindle cell appearance. 18 Calcification is common.

The marked glandular distortion may result in an irregular outline and an impression of a cord‐like growth pattern 11 with extension into fat. This appearance may be mistaken for foci of ILC, particularly when present in specimens that also contain ILC, possibly leading to overestimation of disease extent and incorrect assessment of margin status (Figure 3A ).

Figure 3.

Figure 3

(A) Sclerosing adenosis can mimic ILC on H&E. (B) p63 stain highlights MECs associated with sclerosing adenosis. (C) DCIS involving sclerosing adenosis, associated with lymphocytes, mimicking IC. (D) p63 stain highlights MECs and aids interpretation as DCIS involving sclerosing adenosis.

Involvement of sclerosing adenosis by apocrine change and in situ carcinoma may mimic IC. Colonisation of sclerosing adenosis by florid lobular carcinoma in situ (LCIS) may cause fusion of the affected lobules obscuring surrounding myoepithelium, intervening stroma, and acinar architecture, leading to a low‐magnification appearance of a solid sheet of IC cells. 19 Involvement of sclerosing adenosis by DCIS may also be misinterpreted as IC. This is particularly problematic in cases of extensive DCIS with colonisation of sclerotic lobules and accompanying periglandular lymphocytes, mimicking high‐grade IC (Figure 3C,D ). Perineural infiltration may also be observed in sclerosing adenosis that may lead to misdiagnosis if the pathologist is not aware of this pitfall. 20

Sclerosing adenosis is the benign breast lesion most commonly misdiagnosed as malignant. 5 The most helpful clues to the correct diagnosis are appreciation of the lobulo‐centric arrangement of the affected lobules on H&E at low power, and that although glands may extend into fat, they do not disrupt it. 19 , 21 , 22 , 23 (Figure 3A,B ). IHC demonstration of the peripheral MEC layer is very useful, particularly in excluding a diagnosis of IC in cases of sclerosing adenosis involved by in situ carcinoma, 24 bearing in mind the difficulties that may be encountered due to cell attenuation, as outlined above.

Microglandular adenosis

Microglandular adenosis is a rare infiltrative breast lesion that may mimic tubular carcinoma of the breast. The tubules of microglandular adenosis are typically round, surrounded by basement membrane, and lined by a single layer of epithelial cells that are S100‐positive (Figure 4) and negative for ER, progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). 25 There is no peripheral MEC layer and tubules lie free in stroma and fat with no accompanying stromal reaction. 25 Originally considered to be an entirely benign lesion, microglandular adenosis is now regarded as a possible precursor lesion of triple‐negative breast carcinoma (TNBC). 26 Distinction from tubular carcinoma is important to ensure appropriate treatment. Microglandular adenosis is usually managed by surgical excision without adjuvant therapy. Microglandular adenosis, including atypical forms, shares morphological and molecular features with acinic cell carcinoma of the breast. 27 Both entities lack peripheral MECs and in the context of acinic cell carcinoma, atypical solid clusters resembling atypical microglandular adenosis should be considered part of the acinic cell carcinoma spectrum.

Figure 4.

Figure 4

Microglandular adenosis. (A) Microglandular adenosis has an infiltrative appearance and tubules are seen in the fat (H&E). (B) The tubules contain eosinophilic secretions (H&E). (C) The tubules lack a peripheral MEC layer, in contrast to admixed normal glandular breast tissue (p63 stain). (D) Microglandular adenosis is S100‐positive.

Intraduct papilloma

Benign intraduct papilloma, without accompanying UDH, is rarely confused with encapsulated or solid papillary carcinoma, due to its preserved architecture, well‐developed fibrovascular connective tissue cores covered by epithelial and MECs, and peripheral MEC layer. UDH within an intraduct papilloma is a common finding but is usually focal, often with apocrine metaplasia, and rarely poses diagnostic difficulty. We have seen some examples of intraduct papilloma with exuberant UDH with mitotic activity, colonising the entire lesion with obliteration of the well‐formed papillae to produce a solid appearance that may resemble solid papillary carcinoma or DCIS involving a papilloma (Figure 5). A structured approach to diagnosis that includes defining the nature of the underlying lesion and appreciation of the mixed nature of the epithelial cell proliferation on H&E, supported by IHC, will usually lead to the correct diagnosis.

Figure 5.

Figure 5

(AD) Intraduct papilloma with exuberant UDH may mimic solid papillary carcinoma on H&E. (E,F) The characteristic mosaic staining pattern for CK5/6 and ER demonstrates the polyclonal nature of the epithelial proliferation, supporting the diagnosis of intraduct papilloma with florid UDH.

Intraduct papillomas may become infarcted following CNB, with development of squamous metaplasia that may create an impression of metaplastic carcinoma 28 or, rarely, mucoepidermoid carcinoma. Distortion of residual viable ductal epithelium by fibrosis at the periphery of the lesion may also simulate IC. The low‐power appearance of the lesion, the absence of cytologic features of malignancy, and the presence of peripheral MECs are useful criteria for excluding a diagnosis of IC. 29 Displacement of epithelial cells along the needle tract may occur following CNB and should not be mistaken for carcinoma. 30

Benign breast lesions with myxoid matrix

Fibroadenoma is a common and usually straightforward diagnosis on CNB. On rare occasions, fibroadenomas with a prominent myxoid stroma and sparse glandular elements may create an impression of matrix producing carcinoma. Awareness of this potential pitfall and radiology‐pathology correlation should avert misdiagnosis. 3

Pleomorphic adenoma is a rare occurrence in the breast that may present a confusing appearance on CNB due to the prominent myxoid stroma containing mildly atypical epithelial cells that may express basal type CKs and are hormone receptor‐negative. 6

Spindle cell/mesenchymal lesions

Benign spindle cell lesions (SCLs) of the breast include a variety of conditions, some of which may present an alarming picture histologically, with potential for misdiagnosis as malignant, particularly on CNB. As for all breast lesions, investigation of the clinical history, including duration of symptoms, the radiological findings, and the precise anatomical location of the lesion, in addition to the evaluation of the H&E and IHC findings, may help to avoid error. We advocate grouping SCLs into ‘bland appearing’ and ‘malignant appearing’, which encourages a broad differential diagnosis and consideration of a benign diagnosis when presented with a malignant‐appearing lesion and vice versa, as discussed below (Malignant conditions that may be interpreted as benign). 31

Although relatively uncommon, the benign breast SCLs with potential for misdiagnosis as malignant that are more frequently encountered include scar, fibromatosis, nodular fasciitis, myofibroblastoma, and cellular pseudoangiomatous stromal hyperplasia. Rare benign breast SCLs include inflammatory myofibroblastic tumour, solitary fibrous tumour, spindle cell lipoma, benign smooth muscle, and neural tumours, the spindle cell variant of adenomyoepithelioma, cellular sclerosing adenosis, and myoid hamartoma. 32

Scar

Scar tissue is commonly encountered on CNB, often performed for evaluation of calcification or a mass lesion. Some patients have a history of ipsilateral prior breast surgery or breast trauma. Others may not have any known relevant history, and scar should always be borne in mind when analysing a breast SCL. Recent scar may present as a proliferation of very active fibroblasts with mitotic activity that, in its most florid phase, can be designated as a postoperative spindle cell nodule. 33 Knowledge of any relevant clinical history and a search for accompanying change, e.g. haemosiderin deposition, foamy macrophages, fat necrosis, and foreign body giant cells assist diagnosis (Figure 6). Scar tissue may express smooth muscle actin. CKs, p63, CD34, and β‐catenin are negative.

Figure 6.

Figure 6

(A) Spindle cell proliferation with haemosiderin‐deposition (H&E). (B) Associated fat necrosis (H&E). The features support a diagnosis of scar.

Fibromatosis

Fibromatosis is a rare breast lesion that shows marked morphological similarity to fibromatosis‐like MBC (discussed below). Both lesions tend to have an infiltrative outline with entrapment of peripheral fat and lymphocytic inflammation. The cells of fibromatosis lack cytological atypia and mitoses should be difficult to identify (Figure 7A). This lesion may recur locally but is usually managed by surveillance. 34 , 35

Figure 7.

Figure 7

Fibromatosis. (A) H&E stain shows a cytologically bland spindle cell proliferation. (B) Nuclear β‐catenin staining of fibromatosis.

β‐catenin nuclear staining (Figure 7B) and smooth muscle actin (SMA) cytoplasmic staining are the characteristic IHC findings. CKs, p63, and CD34 are typically negative.

Nodular fasciitis

Nodular fasciitis is also an uncommon benign breast SCL that may present with a history of rapid growth and is usually self‐limiting. The component fibroblasts typically have a stellate configuration with a relatively bland appearance and a feathery growth pattern. The matrix may be myxoid or collagenous and contains extravasated red blood cells (Figure 8). Some lesions are highly cellular, with frequent mitoses that may result in overdiagnosis as malignant. Awareness of the entity, knowledge of the clinical history, and appreciation of the absence of abnormal mitoses will help to avoid this diagnostic pitfall. The cells of nodular fasciitis may be smooth muscle actin‐positive. They are CK and p63 negative. 36

Figure 8.

Figure 8

Nodular fasciitis. H&E stain shows a cellular spindle cell proliferation composed of fibroblasts with a relatively bland appearance associated with some extravasated red blood cells.

Cellular pseudoangiomatous stromal hyperplasia (PASH)

Cellular pseudoangiomatous stromal hyperplasia (PASH) may be mistaken for angiosarcoma due to the interconnecting arrangement of the pseudovascular channels. The channels are lined by relatively bland‐appearing spindle‐shaped myofibroblasts and do not contain red blood cells (Figure 9) that should alert the pathologist to the correct diagnosis. The lining cells are CD34 and PR‐positive and may express actin and desmin. Other vascular markers are negative. 37

Figure 9.

Figure 9

(A,B) Pseudoangiomatous stromal hyperplasia (PASH).

Myofibroblastoma

Myofibroblastoma of the breast may simulate spindle cell MBC. 38 The usually well‐circumscribed nature of the lesion on radiology and the identification of amianthoid fibres and intervening adipocytes on H&E may help the pathologist to reach the correct diagnosis. In addition to ER, myofibroblastoma cells are typically CD34, SMA, and desmin‐positive. CKs and p63 are negative. 36

As mentioned earlier, the epithelioid variant of myofibroblastoma may resemble ILC, 3 with further potential for a false‐positive diagnosis due to ER positivity (Figure 10).

Figure 10.

Figure 10

(A) Epithelioid myofibroblastoma may mimic ILC (H&E). (B) Cells of myofibroblastoma express ER. (C) CD34 stains the myofibroblastoma.

Eosinophilic granular lesions

The differential diagnosis of eosinophilic lesions of the breast, such as histiocytic inflammation and granular cell tumour, includes carcinoma with oncocytic change and apocrine carcinoma. 39

Histiocytic inflammation

Florid histiocytic inflammation may result from a ruptured duct or cyst or develop secondary to a previous breast procedure. Histologically, this may resemble apocrine carcinoma, particularly on CNB where accompanying changes that assist accurate diagnosis, e.g. elements of a duct/cyst or a fibrous reaction, may not be represented (Figure 11). Histiocytes are CD68‐positive and CK‐negative.

Figure 11.

Figure 11

(A) Histiocytic inflammation (H&E), which may resemble histiocytoid carcinoma (H&E) shown in B and C.

Granular cell tumour

Granular cell tumour is a rare, generally benign, breast tumour that usually occurs in women. It has an infiltrative growth pattern and may mimic IC both clinically and histologically. The tumour is composed of round to polygonal cells with abundant, eosinophilic, granular cytoplasm, due to accumulation of lysosomes, that may mimic apocrine carcinoma. Perineural and perivascular involvement may be seen 40 and lesions located in the dermis may be associated with overlying pseudoepitheliomatous hyperplasia, with further potential for misdiagnosis. Clues to the correct diagnosis include appreciation of the small, centrally located, hyperchromatic, nuclei, scarce mitoses and indistinct cell borders. Lesional cells are strongly S100‐positive and are also CD68‐ and NSE‐ positive. 41 CKs, melan‐A and HMB45 are negative, distinguishing this tumour from carcinoma and melanoma 42 (Figure 12).

Figure 12.

Figure 12

Granular cell tumour (H&E).

Others

Epithelial cell displacement following needling procedures

Epithelial cell displacement in breast tissue may occur following diagnostic needling procedures and lead to interpretative difficulties in subsequent excision specimens that have been performed within a short time interval. 43 As noted above, this phenomenon is seen most frequently in association with papillary lesions, 30 , 44 but has also been reported following liposuction for gynaecomastia 45 and following biopsy of radial scar, 46 DCIS, 47 and IC. 48

While displaced epithelial cells are considered biologically insignificant, 49 their presence in breast tissue without a MEC layer at the epithelial stromal interface or within vascular channels may result in overdiagnosis of IC. The presence of epithelial displacement in malignant tumours may affect the assessment of tumour/lesion size and margin status and lead to overdiagnosis of lymphovascular invasion (Figure 13) and IC, particularly in DCIS, with potential management implications. 47 IHC does not assist the distinction of displaced epithelial cells from foci of IC, as they are typically devoid of a MEC layer. The correct interpretation of displaced epithelial cells is facilitated by awareness of a prior needling procedure and appreciation of the accompanying postprocedure histological changes including haemorrhage, granulation tissue, acute and chronic inflammation, and fat necrosis with accumulation of vacuolated and haemosiderin‐laden macrophages. 50 The time interval between the needling procedure and the surgical excision should also be taken into consideration. 48

Figure 13.

Figure 13

(A) Epithelial displacement may mimic IC (H&E). (B) Epithelial displacement in a lymphatic space mimicking lymphovascular invasion in a patient with a benign intraduct papilloma.

In contrast, early invasive foci in encapsulated papillary carcinoma may be accompanied by regressive stromal changes, creating the impression of epithelial displacement. Distinguishing features include the multifocal nature of early invasion with reactive stromal changes. The latter are considered to be due to repeated small ruptures of the cystic papillary structure rather than secondary to a diagnostic needling procedure in which the changes are typically localised and often linear.

Artefactual displacement of DCIS (‘toothpaste effect’)

Artefactual displacement of DCIS (ADDCIS) not associated with needling procedures may occur secondary to mechanical compression of breast specimens leading to overdiagnosis of IC, as recently reported. 51 Shabihkhani et al. identified 16 cases of ADDCIS not associated with needle tract changes, referred for evaluation of, or with a definite diagnosis, of IC. Most occurred in lumpectomy specimens; foci ranged in size from <1 to 5 mm and involved the breast stroma in a non‐lobular distribution. All foci measuring >4 mm demonstrated a linear pattern of displacement, which is a useful diagnostic criterion. Smudged chromatin and the absence of a stromal reaction are also helpful features. Similar to displaced epithelium following needling procedures, MEC IHC is negative. None of the patients in whom follow‐up was available developed metastases, supporting the view that ADDCIS does not represent IC.

Collagenous spherulosis

Collagenous spherulosis is a relatively rare, benign, lesion characterised by deposits of acellular, eosinophilic, or basophilic, matrix surrounded by basement membrane, myoepithelial, and ductal epithelial cells. It is usually an incidental microscopic finding. Collagenous spherulosis may be involved by LCIS producing a ’collision‘ lesion with a cribriform appearance that may be mistaken for low‐grade cribriform DCIS. 52 , 53 , 54

Both lesions are characterised by the combination of small cells with uniform nuclei and round, punched‐out spaces. Identifying eosinophilic basement membrane or basophilic mucoid material within the spaces, observing the absence of polarised epithelial cells around the cribriform spaces, and noting that the proliferation is composed of dyscohesive epithelial cells with intracytoplasmic vacuoles may lead the pathologist to consider this diagnosis.

Awareness of this entity may prompt the pathologist to use IHC to further investigate the biopsy findings and avoid a misdiagnosis of low‐grade DCIS. LCIS cells are typically E‐cadherin‐negative, while those of DCIS are E‐cadherin‐positive. MEC IHC will demonstrate MECs around the spaces of collagenous spherulosis, whereas the cribriform spaces of DCIS are not lined by MECs (Figure 14).

Figure 14.

Figure 14

(A) Collagenous spherulosis colonized by LCIS (H&E). (B) E‐cadherin is negative in LCIS.

Collagenous spherulosis may also mimic adenoid cystic carcinoma due to the presence of basement membrane‐like material. Collagenous spherulosis rarely forms a mass lesion and is typically c‐kit (CD117)‐negative, in contrast to adenoid cystic carcinoma. 55 , 56

Lactactional change

Lactational changes may be florid and with significant nuclear changes including nucleomegaly, pleomorphism, and a hobnail appearance that may lead to a diagnosis of flat or hypersecretory type DCIS (Figure 15). The diagnosis of DCIS should be restricted to cases showing architectural atypia including cribriform, solid or micropapillary growth patterns, and definite central (comedo‐type) necrosis or HER2 IHC positivity (IHC score 3+). 57 , 58 Expression of p53 and/or a high Ki67 score also favour a diagnosis of DCIS.

Figure 15.

Figure 15

Lactational change. (A) Lactational change (H&E). (B,C) Lactational change with a deposit of calcium phosphate.

Hypersecretory thyroid‐like adenosis

Hypersecretory thyroid‐like adenosis is a recently described rare entity that consists of multiple dilated glands of varying sizes, with a preserved lobular architecture, and containing dense eosinophilic diastase PAS‐positive colloid‐like secretions. The glands are lined by ER‐negative epithelial cells and are surrounded by a dense basement membrane. Peripheral MECs are absent. The epithelial cells do not display cytological atypia or mitotic activity, which if present should raise suspicion of IC. Hypersecretory thyroid‐like adenosis is more likely to be observed as an incidental finding rather than presenting as a clinical or radiological abnormality. 59

Ectopic breast tissue in lymph nodes

The presence of ectopic breast tissue in axillary lymph nodes may be misinterpreted as nodal metastases, particularly if the breast tissue shows hyperplastic changes.

Malignant conditions that may be misinterpreted as benign

Some malignant breast lesions may appear deceptively bland, resulting in underdiagnosis as either benign, hyperplastic, or locally infiltrative. Some invasive tumours are also more appropriately categorised as in situ lesions due to their indolent biological behaviour.

Spindle cell DCIS

DCIS with spindle cell morphology is a rare entity that may be mistaken for UDH, particularly if it is the sole pattern without any coexisting traditional type DCIS. 60 This form of DCIS is composed exclusively or predominantly of spindle cells arranged in fascicles, whorls, and solid sheets. The fascicular arrangement of the spindle cells may simulate the ‘streaming’ pattern of UDH (Figure 16). Clues to the correct diagnosis include the solid growth pattern with lack of intercellular spaces and peripheral fenestrations. The cells appear uniform with low‐ to intermediate‐grade nuclear atypia. 61 As for other forms of DCIS, the cells are negative for HMW CKs 5/6 and 14. Most cases are ER‐positive and also show neuroendocrine differentiation. 60

Figure 16.

Figure 16

(A,B) Spindle cell DCIS may mimic UDH on H&E. (C) There is an absence of CK5/6 staining in spindle cell DCIS. (D) Spindle cell DCIS shows diffuse strong nuclear expression of ER.

Low‐grade solid DCIS

The distinction of low‐grade solid DCIS from classical LCIS is clinically important due to the current different approaches to management. Classical LCIS is generally managed by mammographic surveillance with consideration of risk‐reducing endocrine therapy. In contrast, DCIS is usually managed as a direct breast cancer precursor with complete surgical excision, often followed by radiation therapy. The diagnosis may be difficult, particularly on CNB. Both are composed of small uniform cells with monomorphic nuclei. Features suggestive of ductal differentiation include cellular cohesion, well‐defined cell borders, and focal columnar shape of the neoplastic cells. The identification of microacini with polarisation of cells around lumina favours a diagnosis of DCIS. 62 Features in favour of a lobular phenotype include lack of cellular cohesion and the presence of intracytoplasmic vacuoles. E‐cadherin IHC is helpful when presented with this differential diagnosis, bearing in mind some of the difficulties associated with interpretation of this IHC, as discussed below.

Encapsulated papillary carcinoma

Although intraduct papilloma is rarely misdiagnosed as malignant, encapsulated papillary carcinoma may present a ‘benign’ appearance at first glance due to its well circumscribed contour, expansile growth pattern, and prominent papillary architecture. A useful clue at low power is the rather fragmented and disrupted nature of the lesion in comparison to the preserved, intact, character of intraduct papilloma (Figure 17). On closer examination, encapsulated papillary carcinoma generally lacks a peripheral MEC layer and the fibrovascular cores within the lesion are covered by epithelial cells only that are cytologically atypical with mitotic activity. 63 , 64 It is important to appreciate that the fibrovascular cores are lined by a single cell type, often with multiple cell layers, in contrast to the cores of intraduct papilloma that are lined by two cell types. In some lesions, the epithelial cells that are in direct contact with the fibrovascular cores assume a globoid appearance that may mimic MECs producing the so‐called ‘dimorphic pattern’. 65 , 66 IHC may be used to clarify the nature of the cells in doubtful cases. In the absence of frank IC, encapsulated papillary carcinoma is currently classified and staged as an in situ lesion. 39

Figure 17.

Figure 17

(A) Fragments of encapsulated papillary carcinoma from a CNB (H&E). The proliferation shows diffuse strong nuclear expression of ER (B) and is negative for CK5/6 (C), confirming the clonal nature of this papillary lesion.

Solid papillary carcinoma

Solid papillary carcinoma is composed of a proliferation of uniform epithelial cells punctuated by delicate fibrovascular cores. Tumours with a round contour and a pushing border are categorised as in situ with or without a peripheral MEC layer. Tumours are categorised as invasive if the outline is jagged, with stromal desmoplasia and/or the cytology of the epithelial cells is high nuclear grade. 63 , 67 These tumours may show neuroendocrine differentiation and are sometimes associated with mucinous carcinoma. 68 , 69 Well‐circumscribed tumours in which the epithelial cells show low to intermediate nuclear grade with spindling effect may present a bland appearance on H&E and the lesion may be mistaken for intraduct papilloma with florid UDH, the opposite situation to that described earlier. 70 , 71 The correct diagnosis is facilitated by recognition of the uniform nature of the epithelial cells, the delicate fibrovascular network, and the absence of MECs within and usually at the periphery of the nodules. In difficult cases, IHC is helpful in demonstrating the absence of MECs and the clonal nature of the epithelial cell proliferation that is diffusely ER‐positive and HMW CK‐negative. 72

Fibromatosis‐like metaplastic carcinoma (FLMBC)

Fibromatosis‐like metaplastic carcinoma (FLMBC) is easy to misdiagnose as fibromatosis or an alternative benign lesion, particularly on CNB, and is the most important differential diagnosis of a bland‐appearing breast SCL 32 (Figure 18). It is typically composed of spindle cells with minimal nuclear pleomorphism, scarce mitotic activity, and no in situ component. The stroma may be sclerotic or collagenised. A peripheral inflammatory cell infiltrate with lymphoid follicles is seen in some tumours. The association of FLMBC with benign sclerosing lesions further adds to the potential for underdiagnosis, particularly in early lesions. FLMBC has a good prognosis compared with other MBCs, but may recur locally and there have been occasional reports of metastases. 73 Recognition is clearly important, as these tumours are currently managed according the protocol for low‐grade IBC. Most tumours express CKs and p63. As CK positivity may be focal and no individual antibody stains all MBCs, 74 a panel of antibodies against epithelial markers is recommended. 24 β‐catenin nuclear staining, which is a characteristic feature of fibromatosis, is not usually seen in FLMBC. However, focal weak nuclear staining may occasionally be seen 75 and should not deter the pathologist from making a diagnosis of FLMBC. CD34 is negative. Definitive diagnosis may be not possible on CNB, in which case diagnostic excision is recommended.

Figure 18.

Figure 18

(A) Fibromatosis‐like metaplastic breast carcinoma has the appearance of a cytologically bland spindle cell proliferation on H&E. (B) Cytokeratin MNF116 reveals the epithelial nature of the lesion and supports a diagnosis of FLMBC.

Low‐grade angiosarcoma

Angiosarcoma is a rare, but usually aggressive, breast tumour that accounts for ~25% of primary breast sarcomas. 76 Angiosarcoma of the breast may be primary or develop secondary to the effects of radiation therapy and shows varied morphology, ranging from a predominantly vasoformative lesion to a tumour with a predominantly solid growth pattern. Primary angiosarcoma is more frequently composed of anastomosing, well‐formed, small to medium sized vascular channels that dissect through adipose tissue compared with secondary angiosarcoma 77 (Figure 19 ). Low‐grade angiosarcoma is composed of well‐formed vascular channels, lined by endothelial cells with minimal atypia, with potential for misdiagnosis as a benign haemangioma. Unlike haemangioma, the vascular proliferation of angiosarcoma does not display a lobulated configuration, the vessels intercommunicate and traverse normal lobules and adipose tissue.

Figure 19.

Figure 19

(A) Angiosarcoma with a vasoformative growth pattern (H&E). (B) The periphery of the lesion has a more subtle appearance, with delicate thin‐walled vessels infiltrating the fat (H&E).

Malignant conditions that may be incorrectly diagnosed as primary breast carcinoma

Metastases to the breast

Metastases to the breast from extramammary primary malignancies are reported to account for between 0.2 and 1.1% of malignant tumours of the breast and represent the first sign of malignancy in up to one‐third of cases. 78 The incidence appears to be increasing, 79 , 80 likely due to more detailed radiological investigations and histological analysis. The most common sources of metastases to the breast are melanoma, lung, gastric and colorectal adenocarcinoma, renal cell carcinoma, sarcoma, and prostatic carcinoma in males. Malignant mesothelioma, neuroendocrine tumours, and squamous cell carcinoma of the uterine cervix may also involve the breast. 78 The clinical presentation may be very similar to that of primary BC and may be symptomatic or screen‐detected. Recognition and accurate diagnosis are essential to guide appropriate management, including avoidance of unnecessary surgery.

The pathologist may be alerted to the possibility of metastases due to the clinical or radiological findings or the patient's past medical history. The presence of multifocal and/or bilateral lesions, although not uncommon in primary BC, should prompt the pathologist to consider metastases. The absence of DCIS may be a clue that the index lesion is a metastasis, but not all primary BCs display an in situ component at diagnosis. Other microscopic clues of extramammary metastases include unusual morphology, e.g. the melanin pigment of melanoma, the classical “garland‐like” necrosis of colorectal carcinoma or the clear cytoplasm of renal cell carcinoma, or a biomarker profile that is discordant with the histological features, e.g. a triple‐negative low‐grade carcinoma.

Comparison with previous histology and IHC work‐up, including breast markers ER, GATA3, GCDFP‐15, SOX10, and other antibodies directed by morphology and clinical history are very useful in clarifying the diagnosis. Although metastases to the breast are usually triple‐negative, ER expression is not specific to breast and is common in gynaecological malignancies. Similarly, GATA3 is expressed in other tumours, including urothelial carcinoma, squamous cell carcinoma of skin, and malignant mesothelioma. 81 , 82 Despite thorough work‐up, some high‐grade malignancies, e.g. lung and ovary, may be morphologically indistinguishable from high‐grade primary BC and it may be difficult to differentiate between synchronous primary tumours and metastatic disease.

Lymphoma

Lymphomas involving the breast account for ~2% of extranodal lymphomas and for up to 0.5% of all breast malignancies. 83 Patients usually present symptomatically, but breast lymphoma may also be diagnosed on screening mammography. Diffuse large B‐cell lymphoma is the most common subtype in patients with localised disease, followed by extranodal marginal zone lymphoma of mucosal associated lymphoid tissue (MALT lymphoma) and these usually manifest as a solitary breast mass. Follicular lymphoma, mantle cell lymphoma, and small lymphocytic lymphoma are more common in patients who have disseminated disease and may present as multifocal breast lesions. 84 Clinically and radiologically, breast lymphoma presents a very similar picture to primary BC, although a recent study suggests that lymphoma can be distinguished on imaging due to larger lesion size, regular shape, lymph node involvement, and the absence of calcification. 85

Histologically, low‐grade breast lymphomas may mimic ILC due to the uniform, dyscohesive nature of the cell population and the presence of dense collagen, particularly in follicular lymphoma, that may produce a single cell pattern of infiltration (Figure 20). Awareness of this potential pitfall when considering a diagnosis of ILC on CNB, noting the absence of atypical lobular hyperplasia/LCIS and unexpected hormone receptor negativity, should avert misdiagnosis and unnecessary surgery. High‐grade lymphoma may be mistaken for triple‐negative IBC, no special type (NST), grade 3. Careful appraisal of cytological and architectural detail, the absence of DCIS, and a low threshold to further investigate triple‐negative BCs with IHC assist accurate recognition. When evaluating multifocal or bilateral breast lesions, it is always worth considering the possibility of lymphoma or metastatic disease.

Figure 20.

Figure 20

(A,B) Solid lobular carcinoma, low‐power and high‐power images (H&E). (C) Cytokeratin AE1AE3 IHC confirms the epithelial nature of the tumour cells. (D,E) low‐grade lymphoma, low‐power and high‐power images (H&E). (F) CD20 stains the low‐grade lymphoma.

High‐grade angiosarcoma

In contrast to low‐grade angiosarcoma, discussed above, high‐grade angiosarcoma is composed of solid sheets of spindle and epithelioid cells with variable degrees of vasoformation. Mitotic activity, necrosis, and blood lakes are commonly seen. Angiosarcoma with predominant epithelioid morphology may be mistaken for IBC NST, grade 3 on H&E, further compounded by occasional expression of epithelial markers. 86 , 87 Angiosarcoma expresses endothelial markers including CD31 and ERG. These tumours show varying degrees of positivity with antibodies to CD34, D2‐40, and FLi1. 88 Consideration of any prior clinical history and knowledge of the clinical presentation, which may be characteristic, can assist diagnosis. Although a relatively rare occurrence in the breast, most breast pathologists will encounter a number of these tumours during their career and should include angiosarcoma in the differential diagnosis of high‐grade triple‐negative breast tumours.

Others

Some rare BCs may mimic benign lesions or may not be recognised as primary invasive disease. Well‐differentiated forms of adenoid cystic carcinoma may be misclassified as breast cylindroma and bland‐appearing mucoepidermoid carcinoma as intraduct papillary lesion with metaplasia. Some examples of invasive papillary carcinoma may be misinterpreted as papillary DCIS due to the duct‐like profiles of the infiltrative papillary clusters. Some high‐grade ICs may have a well‐circumscribed pushing border with papillary‐like cores, mimicking encapsulated papillary carcinoma. 89 Another pitfall that we have encountered in clinical practice is failing to recognise low‐grade lymphoma in lymph nodes removed for detection of metastatic carcinoma.

Immunohistochemistry pitfalls

E‐cadherin

Accurate diagnosis of ILC is clinically relevant. A biopsy diagnosis of ILC may lead to a request for breast magnetic resonance imaging (MRI) to estimate disease extent, to exclude the possibility of multifocal and bilateral tumours, and to determine the patient’s suitability for breast‐conserving surgery. ILCs tend to have a poorer response to neoadjuvant chemotherapy than IBC NST, 90 which may influence treatment decisions and future clinical guidelines.

The diagnosis of ILC and LCIS is usually based on morphological features. 39 The dyscohesive appearance characteristic of lobular lesions occurs due to underlying loss of E‐cadherin‐mediated cell adhesion. This results in the absence of membranous expression of E‐cadherin, which can be demonstrated using IHC and is frequently used to assist diagnosis in morphologically ambiguous cases (Figure 21A,B). Most ILCs and LCIS (including florid and pleomorphic variants) lack membranous expression of E‐cadherin, whereas IBC NST, DCIS, and normal breast epithelium show strong circumferential membranous staining. However, E‐cadherin expression does not exclude the diagnosis of lobular pathology. It may be present in up to 16% of ILCs 91 and in LCIS 91 , 92 and some high‐grade IBC NSTs may lack E‐cadherin expression. 93

Figure 21.

Figure 21

(A,B) ILC showing loss of E‐cadherin expression. (C,D) ILC showing aberrant E‐cadherin expression with a fragmented pattern of membranous staining, which is decreased relative to background normal ductal epithelium. (E,F) Classical LCIS with negative E‐cadherin. Note also the granular staining of myoepithelial cells.

There is increased awareness of “aberrant” E‐cadherin staining patterns that may be difficult to interpret. 62 , 92 , 94 Rather than being completely E‐cadherin‐negative, ILC and LCIS often exhibit weak or reduced and fragmented membranous expression (Figure 21C,D). Comparison with the strong diffuse membranous staining in adjacent normal breast tissue helps to confirm that the pattern is “aberrant” and supportive of a lobular phenotype. Cytoplasmic expression of E‐cadherin is also described. 92 Some cases with typical classical ILC or LCIS morphology show diffuse strong membranous staining, which may be due to E‐cadherin being inactivated but remaining on the cell surface as a dysfunctional protein. 92 Use of p120 and β‐catenin, if available, may provide evidence of lobular differentiation in these situations. Re‐evaluation of morphology, in conjunction with the E‐cadherin findings, is also helpful.

LCIS with admixed E‐cadherin‐positive benign epithelial cells may cause interpretative difficulty. LCIS may partly involve ducts or acini or undermine existing benign epithelium in a pagetoid pattern. The residual benign native epithelium shows strong expression of E‐cadherin, which may lead to failure to recognise the LCIS process or misinterpretation as DCIS. MECs envelop the acinar and ductal spaces involved by LCIS but may be admixed with LCIS cells. The granular staining of MECs and their cytoplasmic processes (Figure 21E,F) may be mistaken for staining of the LCIS cells. 62 , 95

LCIS cells admixed with UDH may mimic DCIS or atypical ductal hyperplasia. Comparison with H&E helps the pathologist to determine that E‐cadherin positivity is in hyperplastic ductal epithelium, which is usually seen in the central part of the proliferation with a streaming or syncytial appearance. The uniform population of small, round, dyscohesive LCIS cells is predominantly seen at the periphery of the duct. In difficult cases, CK5/6 and ER may also assist diagnosis.

E‐cadherin IHC is not recommended when the diagnosis is obvious based on morphology. Unexpected or aberrant staining patterns should not deter the pathologist from classifying a lesion as lobular if the histological features are typical. 62

Breast markers

In addition to morphological assessment, IHC panels that include “breast lineage” markers (discussed below) and the biomarkers, ER, PR, HER2 may be used in combination with organ/site specific markers, e.g. CDX2 (digestive system), TTF1 (lung), and PAX8 (gynaecological tract), to identify the origin of a tumour. The clinical scenario and tumour location will influence the pathologists' selection of IHC markers when trying to determine if a tumour is a metastasis from a breast primary.

  • GCDFP‐15 and mammaglobin

Gross cystic disease fluid protein 15 (GCDFP‐15) and mammaglobin are cytoplasmic stains. These markers are positive in the majority of ER+ and HER2+ BCs, 96 , 97 , 98 but expression may be variable and is frequently focal. Mammaglobin is a more sensitive stain than GCDFP‐15. 99 , 100 These stains are not as useful in TNBC, in which sensitivities are reported to range from 12 to 16% for GCDFP‐15 and from 16 to 32% for mammaglobin. 96 , 97 , 101 Skin adnexal carcinomas may also express these markers and pathologists should be mindful of this when assessing possible cutaneous recurrences.

  • GATA3

GATA3, a nuclear transcription factor, is involved in the differentiation of breast luminal epithelial cells and is associated with the ER signalling pathway in BC. 102 The sensitivity of GATA3 is strongly associated with ER status. It is expressed in over 90% of ER+ BCs and usually shows strong, diffuse staining. 99 It is a more sensitive marker for ER+ BC than GCDFP‐15 or mammaglobin and the nuclear staining pattern of GATA3 is also easier to interpret. Lymphocytes also express GATA3 and represent a useful internal positive control in many cases. 99 , 102 GATA3 is less sensitive in TNBC, with expression rates of 66%, 67%, and 82% reported in the literature. 96 , 103 , 104 GATA3 is a helpful marker in the investigation of suspected metastatic BC. There is high concordance of GATA3 expression between primary breast tumours and metastases. 103 , 104 Concordance remains high even with loss of ER expression. 103

While GATA3 is a sensitive marker of both breast and urothelial carcinomas, 81 , 82 increasing experience with this marker has shown that it stains other epithelial tumours, including cutaneous basal and squamous cell carcinomas, skin adnexal tumours, choriocarcinoma, chromophobe renal cell carcinoma, salivary gland, and pancreatic ductal adenocarcinomas. Miettinen et al. have also shown that it is expressed with lower frequency (<10%) in a wide variety of carcinomas, e.g. adenocarcinomas of pulmonary, gastrointestinal, and gynaecological origin. 81

  • SOX10

SOX10 is a transcription factor that plays a role in the development of neural crest cells, melanocytes, and Schwann cells 105 and was initially used in the IHC investigation of melanocytic and neural tumours. SOX10 is also expressed in breast and salivary gland MECs 106 and stains a group of breast and salivary gland tumours with MEC and basal cell differentiation. 107 , 108 SOX10 is a valuable IHC marker, as it stains a subset of BCs that do not express GCDFP‐15, mammaglobin, or GATA3. In contrast to GATA3, SOX10 expression is rare in ER+ BC, but it is identified in 31 to 69% of TNBCs, 104 , 107 , 109 , 110 particularly in those that are GATA3‐negative, with reported expression rates of 30–74% in that subgroup. 104 , 109 , 110

The combination of GATA3 and SOX10 may help to elucidate a primary breast origin in suspected TNBC metastases 104 , 109 , 110 and are complementary, as they appear to preferentially stain different subsets; GATA3 staining is seen in metastatic BC from a non‐TNBC primary tumour with loss of biomarker expression, while SOX10 labels metastases from primary TNBC. 104 Similar to GATA3, SOX10 expression is reported to be stable between primary and metastatic tumours. 104

SOX10 also stains primary and metastatic melanoma, which represents a potential pitfall. A similar drawback is seen with S100, another marker of neural and melanocytic differentiation that is expressed in some BCs. To avoid diagnostic error, SOX10 and S100 should be interpreted in conjunction with other markers such as CKs, which can be used to confirm the epithelial nature of the tumour when metastatic TNBC is suspected.

In addition to the expression of breast markers by tumours of non‐breast origin, it is also noteworthy that BC may express markers that are more typically expressed by other tumours, e.g. CDX2 and PAX8, as recently reported. 111 This serves to remind us to evaluate IHC findings in conjunction with H&E morphology and in a clinical and radiological context.

Myoepithelial cell immunohistochemistry

Identification of MECs, both on H&E and on IHC, is an important element in the diagnostic workup and categorisation of benign and malignant glandular, papillary, and intraductal proliferative lesions. The absence of MECs at the periphery of a lesion may support a diagnosis of invasive malignancy but must be interpreted in context and with caution. As outlined earlier, encapsulated and solid papillary carcinoma may be categorised as in situ lesions despite the absence of a peripheral MEC layer.

Where demonstration of MECs is required for categorisation, it is recommended that at least two MEC markers are used, usually a combination of a nuclear and cytoplasmic stain. 24

p63 is a nuclear stain with high sensitivity and specificity for MECs, with no expression in blood vessels or myofibroblasts. Cytoplasmic MEC markers that are commonly used include SMA, calponin, smooth muscle myosin heavy chain (SMMHC), and CD10, which display varying degrees of sensitivity and specificity. 18 The main disadvantage of cytoplasmic MEC markers is their tendency to stain myofibroblasts and blood vessels that can hinder interpretation. This can be particularly problematic in the distinction of distorted ducts involved by DCIS from foci of IC, in which myofibroblasts are aligned at the periphery of the invasive foci mimicking a MEC layer on cytoplasmic MEC IHC (Figure 22).

Figure 22.

Figure 22

Myofibroblasts aligned alongside nests of IC cells with non‐specific calponin staining (calponin IHC). This non‐specific staining contrasts with strong cytoplasmic staining of MECs surrounding normal ductal epithelium and DCIS.

Antibodies to HMW CKs may also be used to stain MECs but may show variation in staining intensity with false‐negative results. In addition to staining myofibroblasts, CD10 may show non‐specific reactivity to epithelial cells. It is reported to be a more specific marker for the identification of MECs in LCIS and low‐grade DCIS than in high‐grade DCIS. 112

Reduced p63 nuclear staining intensity has been reported in sclerosing adenosis, DCIS, and apocrine lesions, which may lead to difficulties in diagnosis. 113 , 114 , 115 This may be explained by MEC hypertrophy with cell elongation and increased internuclear distance making it difficult to visualise the nuclear staining. However, reduced expression of cytoplasmic MEC markers has also been reported in benign and noninvasive breast lesions, 113 , 114 , 115 leading to postulation that reduced, or, even absent, MECs may be caused by immunophenotypic alterations, particularly in apocrine lesions―the ‘missing MECs’. 115 The evaluation of apocrine lesions, in particular, should take account of all cytological and architectural features and categorisation should not be rendered solely on the presence or absence of MECs. 115

Other potential pitfalls include the expression of p63 in cells with squamous or basal differentiation. 2 This may lead to challenges in the diagnosis of some low‐grade TNBCs, e.g. low‐grade adenosquamous carcinoma in which the epithelial cells of small glandular structures show p63 staining that may be mistaken for MECs (Figure 23).

Figure 23.

Figure 23

(AC) Low‐grade adenosquamous carcinoma (H&E). (D) Epithelial cells of the small nests within the low‐grade adenosquamous carcinoma show focal p63 staining.

MECs may also express lymphovascular markers, e.g. D2‐40. The use of a specific MEC marker will help to distinguish between lymphovascular invasion (p63‐negative and D2‐40‐positive) and DCIS with retraction in which the peripheral MECs express both markers.

Conclusion

Knowledge of the potential pitfalls in the evaluation of H&E‐stained sections and IHC preparations and the ability to maintain an open mind in the diagnostic work‐up of operative and non‐operative breast specimens are essential elements of the breast pathologist's skill set to avoid interpretative error and to ensure the safe and correct patient management.

Ethical approval and consent to participate

Not applicable.

Availability of data and materials

Not applicable.

Funding

Not applicable.

Conflict of interest

None.

Acknowledgements

All authors contributed to the writing of this article and to the preparation of images. All have approved the final version. Open access funding provided by IReL.

Data availability statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

References

  • 1. Azzopardi JG. Problems in breast pathology. London: Saunders, 1979. [PubMed] [Google Scholar]
  • 2. Li Z, Dabbs DJ. Avoiding "false positive" and "false negative" immunohistochemical results in breast pathology. Pathobiology 2022; 1‐15; 1–15. [DOI] [PubMed] [Google Scholar]
  • 3. Shaaban AM. Diagnostic pitfalls in needle core biopsy of the breast. Diagn. Histopathol. 2022; 28; 156–160. [Google Scholar]
  • 4. Kim WG, Cummings MC, Lakhani SR. Pitfalls and controversies in pathology impacting breast cancer management. Expert Rev. Anticancer Ther. 2020; 20; 205–219. [DOI] [PubMed] [Google Scholar]
  • 5. Harrison BT, Dillon DA, Richardson AL, Brock JE, Guidi AJ, Lester SC. Quality assurance in breast pathology: lessons learned from a review of amended reports. Arch. Pathol. Lab. Med. 2017; 141; 260–266. [DOI] [PubMed] [Google Scholar]
  • 6. Rakha EA, Aleskandarany MA, Samaka RM, Hodi Z, Lee AH, Ellis IO. Pleomorphic adenoma‐like tumour of the breast. Histopathology 2016; 68; 405–410. [DOI] [PubMed] [Google Scholar]
  • 7. Talwar A, Jain S, Ahuja A, Paliwal P. A rare case of epithelioid Myofibroblastoma of breast mimicking lobular carcinoma on Trucut biopsy: A diagnostic pitfall and literature review. Indian J. Surg. Oncol. 2021; 12(Suppl 1); 201–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Rakha EA, El‐Sayed ME, Reed J, Lee AHS, Evans AJ, Ellis IO. Screen‐detected breast lesions with malignant needle core biopsy diagnoses and no malignancy identified in subsequent surgical excision specimens (potential false‐positive diagnosis). Eur. J. Cancer 2009; 45; 1162–1167. [DOI] [PubMed] [Google Scholar]
  • 9. Rosen PP, Caicco JA. Florid papillomatosis of the nipple. A study of 51 patients, including nine with mammary carcinoma. Am. J. Surg. Pathol. 1986; 10; 87–101. [PubMed] [Google Scholar]
  • 10. Jones MW, Tavassoli FA. Coexistence of nipple duct adenoma and breast carcinoma: a clinicopathologic study of five cases and review of the literature. Mod. Pathol. 1995; 8; 633–636. [PubMed] [Google Scholar]
  • 11. D'Alfonso TM, Shin SJ. Small glandular proliferations of the breast. Surg Pathol Clin. 2012; 5; 591–643. [DOI] [PubMed] [Google Scholar]
  • 12. Rosen PP. Syringomatous adenoma of the nipple. Am. J. Surg. Pathol. 1983; 7; 739–745. [PubMed] [Google Scholar]
  • 13. Boecker W, Junkers T, Reusch M et al. Origin and differentiation of breast nipple syringoma. Sci. Rep. 2012; 2; 226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Ha SM, Cha JH, Shin HJ et al. Radial scars/complex sclerosing lesions of the breast: Radiologic and clinicopathologic correlation. BMC Med. Imaging 2018; 18; 39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Nassar A, Conners AL, Celik B, Jenkins SM, Smith CY, Hieken TJ. Radial scar/complex sclerosing lesions: a clinicopathologic correlation study from a single institution. Ann. Diagn. Pathol. 2015; 19; 24–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Pinder SE, Shaaban A, Deb R et al. NHS breast screening multidisciplinary working group guidelines for the diagnosis and management of breast lesions of uncertain malignant potential on core biopsy (B3 lesions). Clin. Radiol. 2018; 73; 682–692. [DOI] [PubMed] [Google Scholar]
  • 17. Wilsher MJ. Significance of adenosquamous proliferation in breast lesions. J. Clin. Pathol. 2021; 74; 559–567. [DOI] [PubMed] [Google Scholar]
  • 18. Maguire A, Quinn C, Maguire J, Foschini MPF. The role of immunohistochemistry in diagnostic breast pathology. Diagn. Histopathol. 2022; 28; 126–141. [Google Scholar]
  • 19. Chen Y‐Y, Rabban JT. Patterns of lobular carcinoma in situ and their diagnostic mimics in Core needle biopsies. Pathol. Case Rev. 2009; 14; 141–146. [Google Scholar]
  • 20. Gobbi H, Jensen RA, Simpson JF, Olson SJ, Page DL. Atypical ductal hyperplasia and ductal carcinoma in situ of the breast associated with perineural invasion. Hum. Pathol. 2001; 32; 785–790. [DOI] [PubMed] [Google Scholar]
  • 21. Richards D, Ayala AA, Wu Y, Middleton LP. Carcinoma in situ involving sclerosing adenosis on Core biopsy: diagnostic pearls to aid the practicing clinician and avoid overtreatment. Oncol Ther. 2020; 8; 81–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Rasbridge SA, Millis RR. Carcinoma in situ involving sclerosing adenosis: a mimic of invasive breast carcinoma. Histopathology 1995; 27; 269–273. [DOI] [PubMed] [Google Scholar]
  • 23. Fechner RE. Lobular carcinoma in situ in sclerosing adenosis. A potential source of confusion with invasive carcinoma. Am. J. Surg. Pathol. 1981; 5; 233–239. [DOI] [PubMed] [Google Scholar]
  • 24. Lee AH. Use of immunohistochemistry in the diagnosis of problematic breast lesions. J. Clin. Pathol. 2013; 66; 471–477. [DOI] [PubMed] [Google Scholar]
  • 25. Foschini MP, Eusebi V. Microglandular adenosis of the breast: a deceptive and still mysterious benign lesion. Hum. Pathol. 2018; 82; 1–9. [DOI] [PubMed] [Google Scholar]
  • 26. Geyer FC, Pareja F, Weigelt B et al. The Spectrum of triple‐negative breast disease: high‐ and low‐grade lesions. Am. J. Pathol. 2017; 187; 2139–2151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Geyer FC, Berman SH, Marchio C et al. Genetic analysis of microglandular adenosis and acinic cell carcinomas of the breast provides evidence for the existence of a low‐grade triple‐negative breast neoplasia family. Mod. Pathol. 2017; 30; 69–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Flint A, Oberman HA. Infarction and squamous metaplasia of intraductal papilloma: a benign breast lesion that may simulate carcinoma. Hum. Pathol. 1984; 15; 764–767. [DOI] [PubMed] [Google Scholar]
  • 29. Tse GM, Ni YB, Tsang JY et al. Immunohistochemistry in the diagnosis of papillary lesions of the breast. Histopathology 2014; 65; 839–853. [DOI] [PubMed] [Google Scholar]
  • 30. Nagi C, Bleiweiss I, Jaffer S. Epithelial displacement in breast lesions: a papillary phenomenon. Arch. Pathol. Lab. Med. 2005; 129; 1465–1469. [DOI] [PubMed] [Google Scholar]
  • 31. Rakha EA, Brogi E, Castellano I, Quinn C. Spindle cell lesions of the breast: a diagnostic approach. Virchows Arch. 2022; 480; 127–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Tay TKY, Tan PH. Spindle cell lesions of the breast - An approach to diagnosis. Semin Diagn Pathol. 2017; 34; 400–409. [DOI] [PubMed] [Google Scholar]
  • 33. Garijo MF, Val‐Bernal JF, Vega A, Val D. Postoperative spindle cell nodule of the breast: pseudosarcomatous myofibroblastic proliferation following endo‐surgery. Pathol. Int. 2008; 58; 787–791. [DOI] [PubMed] [Google Scholar]
  • 34. Duazo‐Cassin L, Le Guellec S, Lusque A et al. Breast desmoid tumor management in France: toward a new strategy. Breast Cancer Res. Treat. 2019; 176; 329–335. [DOI] [PubMed] [Google Scholar]
  • 35. Boland MR, Nugent T, Nolan J et al. Fibromatosis of the breast: a 10‐year multi‐institutional experience and review of the literature. Breast Cancer 2021; 28; 168–174. [DOI] [PubMed] [Google Scholar]
  • 36. Rakha EA, Aleskandarany MA, Lee AH, Ellis IO. An approach to the diagnosis of spindle cell lesions of the breast. Histopathology 2016; 68; 33–44. [DOI] [PubMed] [Google Scholar]
  • 37. Ibrahim RE, Sciotto CG, Weidner N. Pseudoangiomatous hyperplasia of mammary stroma. Some observations regarding its clinicopathologic spectrum. Cancer 1989; 63; 1154–1160. [DOI] [PubMed] [Google Scholar]
  • 38. Wargotz ES, Weiss SW, Norris HJ. Myofibroblastoma of the breast. Sixteen cases of a distinctive benign mesenchymal tumor. Am. J. Surg. Pathol. 1987; 11; 493–502. [DOI] [PubMed] [Google Scholar]
  • 39. Provenzano E, Vranic S, Gatalica ZG. Carcinoma with apocrine differentiation. In World Health Organization classification of tumours of the breast. 5th ed. Lyon: International Agency for Research on Cancer (IARC), 2019. [Google Scholar]
  • 40. Battistella M, Cribier B, Feugeas JP et al. Vascular invasion and other invasive features in granular cell tumours of the skin: a multicentre study of 119 cases. J. Clin. Pathol. 2014; 67; 19–25. [DOI] [PubMed] [Google Scholar]
  • 41. Brown AC, Audisio RA, Regitnig P. Granular cell tumour of the breast. Surg. Oncol. 2011; 20; 97–105. [DOI] [PubMed] [Google Scholar]
  • 42. Gleason BC, Nascimento AF. HMB‐45 and Melan‐a are useful in the differential diagnosis between granular cell tumor and malignant melanoma. Am. J. Dermatopathol. 2007; 29; 22–27. [DOI] [PubMed] [Google Scholar]
  • 43. Youngson BJ, Cranor M, Rosen PP. Epithelial displacement in surgical breast specimens following needling procedures. Am. J. Surg. Pathol. 1994; 18; 896–903. [DOI] [PubMed] [Google Scholar]
  • 44. Douglas‐Jones AG, Verghese A. Diagnostic difficulty arising from displaced epithelium after core biopsy in intracystic papillary lesions of the breast. J. Clin. Pathol. 2002; 55; 780–783. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. McLaughlin CS, Petrey C, Grant S, Ransdell JS, Reynolds C. Displaced epithelium after liposuction for gynecomastia. Int. J. Surg. Pathol. 2011; 19; 510–513. [DOI] [PubMed] [Google Scholar]
  • 46. Bilous M. Breast core needle biopsy: issues and controversies. Mod. Pathol. 2010; 23(Suppl 2); S36–S45. [DOI] [PubMed] [Google Scholar]
  • 47. Koo JS, Jung WH, Kim H. Epithelial displacement into the lymphovascular space can be seen in breast core needle biopsy specimens. Am. J. Clin. Pathol. 2010; 133; 781–787. [DOI] [PubMed] [Google Scholar]
  • 48. Phelan S, O'Doherty A, Hill A, Quinn CM. Epithelial displacement during breast needle core biopsy causes diagnostic difficulties in subsequent surgical excision specimens. J. Clin. Pathol. 2007; 60; 373–376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Tavassoli FA, Pestaner JP. Pseudoinvasion in intraductal carcinoma. Mod. Pathol. 1995; 8; 380–383. [PubMed] [Google Scholar]
  • 50. Tardivon AA, Guinebretière JM, Dromain C, Deghaye M, Caillet H, Georgin V. Histological findings in surgical specimens after core biopsy of the breast. Eur. J. Radiol. 2002; 42; 40–51. [DOI] [PubMed] [Google Scholar]
  • 51. Shabihkhani M, Simpson JF, White MJ, Cimino‐Mathews A, Argani P. Artifactual displacement of ductal carcinoma in situ (ADDCIS) (toothpaste effect): a mimicker of invasive ductal carcinoma. Am. J. Surg. Pathol. 2020; 44; 120–128. [DOI] [PubMed] [Google Scholar]
  • 52. Resetkova E, Albarracin C, Sneige N. Collagenous spherulosis of breast: morphologic study of 59 cases and review of the literature. Am. J. Surg. Pathol. 2006; 30; 20–27. [DOI] [PubMed] [Google Scholar]
  • 53. Eisenberg RE, Hoda SA. Lobular carcinoma in situ with collagenous spherulosis: clinicopathologic characteristics of 38 cases. Breast J. 2014; 20; 440–441. [DOI] [PubMed] [Google Scholar]
  • 54. Sgroi D, Koerner FC. Involvement of collagenous spherulosis by lobular carcinoma in situ. Potential confusion with cribriform ductal carcinoma in situ. Am. J. Surg. Pathol. 1995; 19; 1366–1370. [DOI] [PubMed] [Google Scholar]
  • 55. Rabban JT, Swain RS, Zaloudek CJ, Chase DR, Chen YY. Immunophenotypic overlap between adenoid cystic carcinoma and collagenous spherulosis of the breast: potential diagnostic pitfalls using myoepithelial markers. Mod. Pathol. 2006; 19; 1351–1357. [DOI] [PubMed] [Google Scholar]
  • 56. Cserni G, Quinn CM, Foschini MP et al. Triple‐negative breast cancer histological subtypes with a Favourable prognosis. Cancers (Basel). 2021; 13; 5694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Skalova A, Ryska A, Kajo K, Di Palma S, Kinkor Z, Michal M. Cystic hypersecretory carcinoma: rare and poorly recognized variant of intraductal carcinoma of the breast. Report of five cases. Histopathology 2005; 46; 43–49. [DOI] [PubMed] [Google Scholar]
  • 58. D'Alfonso TM, Ginter PS, Liu YF, Shin SJ. Cystic hypersecretory (in situ) carcinoma of the breast: a clinicopathologic and immunohistochemical characterization of 10 cases with clinical follow‐up. Am. J. Surg. Pathol. 2014; 38; 45–53. [DOI] [PubMed] [Google Scholar]
  • 59. Hung T, Munday WR, Hayes MM. Hypersecretory thyroid‐like adenosis of the breast: a distinct benign entity devoid of myoepithelial cells. J. Clin. Experiment. Pathol. 2015; 5; 209. [Google Scholar]
  • 60. Tan PH, Lui GG, Chiang G, Yap WM, Poh WT, Bay BH. Ductal carcinoma in situ with spindle cells: a potential diagnostic pitfall in the evaluation of breast lesions. Histopathology 2004; 45; 343–351. [DOI] [PubMed] [Google Scholar]
  • 61. Farshid G, Moinfar F, Meredith DJ, Peiterse S, Tavassoli FA. Spindle cell ductal carcinoma in situ. An unusual variant of ductal intra‐epithelial neoplasia that simulates ductal hyperplasia or a myoepithelial proliferation. Virchows Arch. 2001; 439; 70–77. [DOI] [PubMed] [Google Scholar]
  • 62. Canas‐Marques R, Schnitt SJ. E‐cadherin immunohistochemistry in breast pathology: uses and pitfalls. Histopathology 2016; 68; 57–69. [DOI] [PubMed] [Google Scholar]
  • 63. Kulka J, Madaras L, Floris G, Lax SF. Papillary lesions of the breast. Virchows Arch. 2022; 480; 65–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Hill CB, Yeh IT. Myoepithelial cell staining patterns of papillary breast lesions: from intraductal papillomas to invasive papillary carcinomas. Am. J. Clin. Pathol. 2005; 123; 36–44. [DOI] [PubMed] [Google Scholar]
  • 65. Collins LC, Schnitt SJ. Papillary lesions of the breast: selected diagnostic and management issues. Histopathology 2008; 52; 20–29. [DOI] [PubMed] [Google Scholar]
  • 66. Lefkowitz M, Lefkowitz W, Wargotz ES. Intraductal (intracystic) papillary carcinoma of the breast and its variants: a clinicopathological study of 77 cases. Hum. Pathol. 1994; 25; 802–809. [DOI] [PubMed] [Google Scholar]
  • 67. MacGrogan G, Collins LC, Rakha EA, Tan BY, Lerwill M. Solid papillary carcinoma, in situ and invasive. In World Health Organization classification of tumours of the breast. 5th ed. Lyon: International Agency for Research on Cancer (IARC), 2019. [Google Scholar]
  • 68. Quinn CM, Ostrowski JL. Solid papillary carcinoma of the breast. Am. J. Surg. Pathol. 1997; 21; 256–258. [DOI] [PubMed] [Google Scholar]
  • 69. Maluf HM, Koerner FC. Solid papillary carcinoma of the breast. A form of intraductal carcinoma with endocrine differentiation frequently associated with mucinous carcinoma. Am. J. Surg. Pathol. 1995; 19; 1237–1244. [DOI] [PubMed] [Google Scholar]
  • 70. Tay TKY, Tan PH. Papillary neoplasms of the breast‐reviewing the spectrum. Mod. Pathol. 2021; 34; 1044–1061. [DOI] [PubMed] [Google Scholar]
  • 71. Rabban JT, Koerner FC, Lerwill MF. Solid papillary ductal carcinoma in situ versus usual ductal hyperplasia in the breast: a potentially difficult distinction resolved by cytokeratin 5/6. Hum. Pathol. 2006; 37; 787–793. [DOI] [PubMed] [Google Scholar]
  • 72. Maeda I, Tajima S, Kanemaki Y, Tsugawa K, Takagi M. Use of immunohistochemical analysis of CK5/6, CK14, and CK34betaE12 in the differential diagnosis of solid papillary carcinoma in situ from intraductal papilloma with usual ductal hyperplasia of the breast. SAGE Open Med. 2018; 6; 2050312118811542. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73. Victoor J, Bourgain C, Vander Borght S, Vanden Bempt I, De Rop C, Floris G. Fibromatosis‐like metaplastic carcinoma: a case report and review of the literature. Diagn. Pathol. 2020; 15; 20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74. Rakha EA, Coimbra ND, Hodi Z, Juneinah E, Ellis IO, Lee AH. Immunoprofile of metaplastic carcinomas of the breast. Histopathology 2017; 70; 975–985. [DOI] [PubMed] [Google Scholar]
  • 75. Lacroix‐Triki M, Geyer FC, Lambros MB et al. Beta‐catenin/Wnt signalling pathway in fibromatosis, metaplastic carcinomas and phyllodes tumours of the breast. Mod. Pathol. 2010; 23; 1438–1448. [DOI] [PubMed] [Google Scholar]
  • 76. Adem C, Reynolds C, Ingle JN, Nascimento AG. Primary breast sarcoma: clinicopathologic series from the Mayo Clinic and review of the literature. Br. J. Cancer 2004; 91; 237–241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77. Antonescu C. Malignant vascular tumors‐‐an update. Mod. Pathol. 2014; 27(Suppl 1); S30–S38. [DOI] [PubMed] [Google Scholar]
  • 78. Kulka J, Varga Z. Metastases to the breast. In World Health Organization classification of tumours of the breast. 5th ed. Lyon: International Agency for Research on Cancer (IARC), 2019. [Google Scholar]
  • 79. Lee AHS, Hodi Z, Soomro I et al. Histological clues to the diagnosis of metastasis to the breast from extramammary malignancies. Histopathology 2020; 77; 303–313. [DOI] [PubMed] [Google Scholar]
  • 80. Lee AH. The histological diagnosis of metastases to the breast from extramammary malignancies. J. Clin. Pathol. 2007; 60; 1333–1341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81. Miettinen M, McCue PA, Sarlomo‐Rikala M et al. GATA3: a multispecific but potentially useful marker in surgical pathology: a systematic analysis of 2500 epithelial and nonepithelial tumors. Am. J. Surg. Pathol. 2014; 38; 13–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82. Liu H, Shi J, Wilkerson ML, Lin F. Immunohistochemical evaluation of GATA3 expression in tumors and normal tissues: a useful immunomarker for breast and urothelial carcinomas. Am. J. Clin. Pathol. 2012; 138; 57–64. [DOI] [PubMed] [Google Scholar]
  • 83. Picasso R, Tagliafico A, Calabrese M et al. Primary and secondary breast lymphoma: focus on epidemiology and imaging features. Pathol. Oncol. Res. 2020; 26; 1483–1488. [DOI] [PubMed] [Google Scholar]
  • 84. Talwalkar SS, Miranda RN, Valbuena JR, Routbort MJ, Martin AW, Medeiros LJ. Lymphomas involving the breast: a study of 106 cases comparing localized and disseminated neoplasms. Am. J. Surg. Pathol. 2008; 32; 1299–1309. [DOI] [PubMed] [Google Scholar]
  • 85. Li Y, Zhang Y, Wang W, Wei C, Zhao D, Zhang W. A comprehensive model for diagnosis of primary breast lymphoma differentiated from breast cancer and prognosis evaluation of surgical treatment. Front. Oncol. 2022; 12; 858696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 86. Al‐Abbadi MA, Almasri NM, Al‐Quran S, Wilkinson EJ. Cytokeratin and epithelial membrane antigen expression in angiosarcomas: an immunohistochemical study of 33 cases. Arch. Pathol. Lab. Med. 2007; 131; 288–292. [DOI] [PubMed] [Google Scholar]
  • 87. Fletcher CD, Beham A, Bekir S, Clarke AM, Marley NJ. Epithelioid angiosarcoma of deep soft tissue: a distinctive tumor readily mistaken for an epithelial neoplasm. Am. J. Surg. Pathol. 1991; 15; 915–924. [DOI] [PubMed] [Google Scholar]
  • 88. Folpe AL, Chand EM, Goldblum JR, Weiss SW. Expression of Fli‐1, a nuclear transcription factor, distinguishes vascular neoplasms from potential mimics. Am. J. Surg. Pathol. 2001; 25; 1061–1066. [DOI] [PubMed] [Google Scholar]
  • 89. Rakha EA, Varga Z, Elsheik S, Ellis IO. High‐grade encapsulated papillary carcinoma of the breast: an under‐recognized entity. Histopathology 2015; 66; 740–746. [DOI] [PubMed] [Google Scholar]
  • 90. O'Connor DJ, Davey MG, Barkley LR, Kerin MJ. Differences in sensitivity to neoadjuvant chemotherapy among invasive lobular and ductal carcinoma of the breast and implications on surgery‐a systematic review and meta‐analysis. Breast 2022; 61; 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 91. Rakha EA, Patel A, Powe DG et al. Clinical and biological significance of E‐cadherin protein expression in invasive lobular carcinoma of the breast. Am. J. Surg. Pathol. 2010; 34; 1472–1479. [DOI] [PubMed] [Google Scholar]
  • 92. Ginter PS, D'Alfonso TM. Current concepts in diagnosis, molecular features, and Management of Lobular Carcinoma in Situ of the breast with a discussion of morphologic variants. Arch. Pathol. Lab. Med. 2017; 141; 1668–1678. [DOI] [PubMed] [Google Scholar]
  • 93. Rakha EA, Abd El Rehim D, Pinder SE, Lewis SA, Ellis IO. E‐cadherin expression in invasive non‐lobular carcinoma of the breast and its prognostic significance. Histopathology 2005; 46; 685–693. [DOI] [PubMed] [Google Scholar]
  • 94. Dabbs DJ, Schnitt SJ, Geyer FC et al. Lobular neoplasia of the breast revisited with emphasis on the role of E‐cadherin immunohistochemistry. Am. J. Surg. Pathol. 2013; 37; e1–e11. [DOI] [PubMed] [Google Scholar]
  • 95. Dabbs DJ, Bhargava R, Chivukula M. Lobular versus ductal breast neoplasms: the diagnostic utility of p120 catenin. Am. J. Surg. Pathol. 2007; 31; 427–437. [DOI] [PubMed] [Google Scholar]
  • 96. Krings G, Nystrom M, Mehdi I, Vohra P, Chen YY. Diagnostic utility and sensitivities of GATA3 antibodies in triple‐negative breast cancer. Hum. Pathol. 2014; 45; 2225–2232. [DOI] [PubMed] [Google Scholar]
  • 97. Luo MH, Huang YH, Ni YB et al. Expression of mammaglobin and gross cystic disease fluid protein‐15 in breast carcinomas. Hum. Pathol. 2013; 44; 1241–1250. [DOI] [PubMed] [Google Scholar]
  • 98. Bhargava R, Beriwal S, Dabbs DJ. Mammaglobin vs GCDFP‐15: an immunohistologic validation survey for sensitivity and specificity. Am. J. Clin. Pathol. 2007; 127; 103–113. [DOI] [PubMed] [Google Scholar]
  • 99. Cimino‐Mathews A. Novel uses of immunohistochemistry in breast pathology: interpretation and pitfalls. Mod. Pathol. 2021; 34(Suppl 1); 62–77. [DOI] [PubMed] [Google Scholar]
  • 100. Gown AM, Fulton RS, Kandalaft PL. Markers of metastatic carcinoma of breast origin. Histopathology 2016; 68; 86–95. [DOI] [PubMed] [Google Scholar]
  • 101. Huo L, Gong Y, Guo M et al. GATA‐binding protein 3 enhances the utility of gross cystic disease fluid protein‐15 and mammaglobin a in triple‐negative breast cancer by immunohistochemistry. Histopathology 2015; 67; 245–254. [DOI] [PubMed] [Google Scholar]
  • 102. Asch‐Kendrick R, Cimino‐Mathews A. The role of GATA3 in breast carcinomas: a review. Hum. Pathol. 2016; 48; 37–47. [DOI] [PubMed] [Google Scholar]
  • 103. Cimino‐Mathews A, Subhawong AP, Illei PB et al. GATA3 expression in breast carcinoma: utility in triple‐negative, sarcomatoid, and metastatic carcinomas. Hum. Pathol. 2013; 44; 1341–1349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 104. Tozbikian GH, Zynger DL. A combination of GATA3 and SOX10 is useful for the diagnosis of metastatic triple‐negative breast cancer. Hum. Pathol. 2019; 85; 221–227. [DOI] [PubMed] [Google Scholar]
  • 105. Mollaaghababa R, Pavan WJ. The importance of having your SOX on: role of SOX10 in the development of neural crest‐derived melanocytes and glia. Oncogene 2003; 22; 3024–3034. [DOI] [PubMed] [Google Scholar]
  • 106. Nonaka D, Chiriboga L, Rubin BP. Sox10: a pan‐schwannian and melanocytic marker. Am. J. Surg. Pathol. 2008; 32; 1291–1298. [DOI] [PubMed] [Google Scholar]
  • 107. Cimino‐Mathews A, Subhawong AP, Elwood H et al. Neural crest transcription factor Sox10 is preferentially expressed in triple‐negative and metaplastic breast carcinomas. Hum. Pathol. 2013; 44; 959–965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 108. Miettinen M, McCue PA, Sarlomo‐Rikala M et al. Sox10‐‐a marker for not only schwannian and melanocytic neoplasms but also myoepithelial cell tumors of soft tissue: a systematic analysis of 5134 tumors. Am. J. Surg. Pathol. 2015; 39; 826–835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 109. Aphivatanasiri C, Li J, Chan R et al. Combined SOX10 GATA3 is most sensitive in detecting primary and metastatic breast cancers: a comparative study of breast markers in multiple tumors. Breast Cancer Res. Treat. 2020; 184; 11–21. [DOI] [PubMed] [Google Scholar]
  • 110. Sejben A, Voros A, Golan A, Zombori T, Cserni G. The added value of SOX10 immunohistochemistry to other breast markers in identifying cytokeratin 5‐positive triple‐negative breast cancers as of mammary origin. Pathobiology 2021; 88; 228–233. [DOI] [PubMed] [Google Scholar]
  • 111. Shen T, Zhao J, Zhao M et al. Unusual staining of immunohistochemical markers PAX8 and CDX2 in breast carcinoma: a potential diagnostic pitfall. Hum. Pathol. 2022; 125; 35–47. [DOI] [PubMed] [Google Scholar]
  • 112. Shousha S, Forbes G, Hopkins I, Wright JA. CD10‐positive myoepithelial cells are usually prominent around in situ lobular neoplasia of the breast and much less prominent or absent in DCIS. J. Clin. Pathol. 2016; 69; 702–705. [DOI] [PubMed] [Google Scholar]
  • 113. Hilson JB, Schnitt SJ, Collins LC. Phenotypic alterations in myoepithelial cells associated with benign sclerosing lesions of the breast. Am. J. Surg. Pathol. 2010; 34; 896–900. [DOI] [PubMed] [Google Scholar]
  • 114. Hilson JB, Schnitt SJ, Collins LC. Phenotypic alterations in ductal carcinoma in situ‐associated myoepithelial cells: biologic and diagnostic implications. Am. J. Surg. Pathol. 2009; 33; 227–232. [DOI] [PubMed] [Google Scholar]
  • 115. Tramm T, Kim JY, Tavassoli FA. Diminished number or complete loss of myoepithelial cells associated with metaplastic and neoplastic apocrine lesions of the breast. Am. J. Surg. Pathol. 2011; 35; 202–211. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Not applicable.

Data sharing is not applicable to this article as no new data were created or analyzed in this study.


Articles from Histopathology are provided here courtesy of Wiley

RESOURCES