A proposal for the histopathological diagnosis of ductal carcinoma in situ of the Breast

BACKGROUND: As the incidence of ductal carcinoma in situ (DCIS) is increasing, it is necessary to make a guideline for the pathological examination and diagnosis of DCIS, by creating criteria based on clinical and biological aspects of the disease. METHOD: We collected biopsy specimens originally diagnosed as benign lesions, from patients who subsequently developed invasive carcinoma in the ipsilateral breast. The histology of the biopsy specimens was re-evaluated principally according to the 1995 Philadelphia Consensus on DCIS. Histopathological agreement on each biopsy specimen was made by the JBCS Study Group members under a multiviewer microscope. In the course of making conclusive agreements among the pathologists, we developed a consensus for the histopathological diagnosis of DCIS, especially non-comedo types. RESULTS: DCIS is defined as a carcinoma of ductal epithelial origin, without any evidence of stromal invasion. It is necessary to note the methods of pathologic examination required to diagnose DCIS. Stromal invasion is an important prognostic factor, and should be diagnosed with caution. Classification of proliferative ductal lesions as benign or malignant (DCIS), the subtype of DCIS (nuclear grade, architecture, and necrosis), and the histological grading of DCIS are proposed and recommended. CONCLUSION: Although we have made a new proposal according to current concepts, there are still several unresolved problems. Thus further examination and modification will be necessary in the future.     

Histological subtypes of ductal carcinoma<Emphasis Type="Italic">in situ</Emphasis> of the breast

BACKGROUND: It has become common to divide ductal carcinoma in situ (DCIS) of the breast into two main groups, comedo or noncomedo by tumor morphology. But noncomedo DCIS can be further stratified into several morphological patterns that exhibit several different growth patterns and most DCIS lesions have more than one pattern. In this study, DCIS elements were classified by morphological pattern and the association between predominant or recessive elements of DCIS lesions and clinicopathological findings was evaluated. METHODS: DCIS lesions from 46 patients were studied regarding the histological architectural patterns: comedo, cribriform, papillary, solid and micropapillary. The predominant architectural pattern which comprised more than 50% of the cells of the cancerous lesion was defined as the major element of the tumor and minor elements consisted of less than 50% but more than 25% of cells comprising the tumor. RESULTS: Of 24 tumors containing a comedo pattern as the major or minor element, 9 (38%) had microscopic intraductal spread over 2 cm and 11 (46%) had involvement of lobules, which was significantly higher than that observed in other types. Of 20 tumors containing a cribriform pattern as the major or minor element, 8 (40%) had microscopic intraductal spread over 2 cm and 9 (45%) had involvement of lobules, which was significantly higher than that seen in other types. Of 10 tumors containing a papillary type as the major or minor element, 5 (50%) had discrete multicentric lesions in the ipsilateral or contralateral breast, which was significantly higher than that seen in other types. CONCLUSIONS: DCIS lesions containing a comedo or cribriform element are more likely to have microscopic spread and involvement of lobules and DCIS lesions containing a papillary element are likely to be multicentric, whether the pattern are predominant in the tumor or not.     

p63 expression in normal,hyperplastic and malignant breast tissues

BACKGROUND: p63 is a homologue of the p53 tumor suppressor gene and its protein is selectively expressed in the basal cells of a variety of epithelial tissues. It has recently been confirmed that p63 is expressed in the basal cells of normal prostate glands but not in prostatic carcinomas. Whether expression of p63 in breast correlates with tumor progression is the focus of this study. METHODS: Forty cases, which all contained normal breast tissue, ductal hyperplasia, ductal carcinoma in situ and invasive ductal carcinoma in the same patient were included in this investigation using an indirect immunohistochemical method and double staining. RESULTS: p63 was exclusively expressed in the myoepithelial cells of normal breast, partially expressed in ductal hyperplasia, rarely expressed in carcinoma in situ and not expressed in invasive carcinomas. CONCLUSION: The results suggest an association between loss of p63 expression and progression of breast ductal carcinoma. p63 immunostaining might be of assistance for distinguishing invasive ductal carcinoma from ductal carcinoma in situ or rare questionable ductal hyperplastic lesions, leading to correct therapy clinically.     

Diagnosis of ductal carcinoma in situ (DCIS) and intraductal papilloma using fluorescence in Situ Hybridization (FISH) analysis

BACKGROUND: It is often difficult to pre-operatively diagnose ductal carcinoma in situ (DCIS)or intraductal papilloma (IDP). Current reports show that breast cancer frequently has numerical aberrations of chromosomes 1, 11 and 17. We investigated whether fluorescence in situ hybridization (FISH) analysis using three centromere-specific probes for chromosomes 1, 11 and 17 was feasible for diagnosing intraductal breast lesions. METHODS: Fine-needle aspiration specimens from 102 breast lesions including DCIS (10), invasive ductal carcinoma (IDC) (78), IDP (7), fibroadenoma (6) and mastopathy (1) were examined for numerical aberrations on chromosomes 1, 11, 17 using FISH. If over 15% of all cells showed one signal, the sample was judged monosomic. If over 20% of cells showed three or more signals, it was considered polysomic. If the specimen had an aberration of at least one chromosome, it was judged positive. RESULTS: Nine of 10 DCISs showed numerical aberrations of at least one chromosome whereas 65 of 78 IDCs and 2 of 14 benign lesions (containing 7 IDPs of which one case was positive) showed numerical aberrations on these chromosomes. The proportion of positive results was highest with DCIS. Moreover 6 out of 7 DCISs showed an aberration of all three chromosomes simultaneously and one case showed an aberration of two chromosomes. All aberrations in case of DCIS were polysomic while two benign lesions and 15 IDCs showed a monosomic pattern. CONCLUSION: FISH may enable more accurate diagnosis of intraductal breast lesions.     

Ductal carcinoma<Emphasis Type="Italic">in situ</Emphasis> and related lesions of the breast: Recent advances in pathology practice

The incidence of ductal carcinoma in situ (DCIS) of the breast has increased significantly in Japanese women. It comprises 14.1% (172/1216) of all primary breast cancers at our institute, and nowadays this histological type is familiar to the surgeons and pathologists of any institute. Several subclassifications have been published recently. Most based on nuclear atypia and the presence of comedonecrosis, and sometimes on the structures of the involved glands. These classifications are correlated with the biological behavior, tumor extent and the risk for local recurrences. The diagnostic accuracy of minimally invasive procedures (aspiration biopsy cytology/core needle biopsy) may differ between subclasses. Atypical ductal hyperplasia (ADH) and microinvasive ductal carcinomas are lesions which resemble but deviate from the DCIS spectrum. The incidence of ADH seems to be lower than in Western countries. Patients with ADH may have a risk for subsequent breast cancer, because ADH is frequently associated with contralateral breast carcinomas. Microinvasion should be treated with caution, but we could not find any metastatic foci in microinvasive ductal carcinomas (T1mic). Tentatively, ADH may be treated similarly to non-comedo (low-grade) DCIS cases, according to our limited clinical experience.     

Treatment of noninvasive carcinoma: Fifteen-Year results at the National Cancer Center Hospital in Tokyo

The introduction of screening mammography (MMG) will lead to increased detection of preclinical early breast cancer in Japan. It has become more important to understand the nature of these lesions. We tried to elucidate the long term prognosis and clinical and pathological characteristics of noninvasive cancers. A total of 336 (5.4%) ductal carcinoma in situ (DCIS) and 32 (0.5%) lobular carcinoma in situ (LCIS) were diagnosed in 6 277 breast carcinomas at the National Cancer Center Hospital from 1962 to 1995. Most (80%) LCIS occurred in premenopausal women. LCIS has significantly higher bilaterality than that of DCIS. Local recurrence occurred in approximately 10% of patients after breast conserving surgery for DCIS and LCIS. Four patients died of breast carcinoma, which were initially diagnosed as noninfiltrating carcinoma. The 15-year cause specific survival rates of patients with DCIS and LCIS were 98.5 % and 100 %, respectively.     

Ductal Carcinomain-situ of the breast detected by [F-18] fluorodeoxyglucose positron emission tomography

A 48-year-old Japanese woman underwent [F-18] fluorodeoxyglucose positron emission tomography (FDG-PET) as part of her medical examination. A small hot spot was detected in her right breast. Quadrantectomy with sentinel lymph node (SN) biopsy using an endoscope was performed, and ductal carcinoma in-situ of the breast was diagnosed. The tumor size was 0.9 cm in its greatest diameter, and there were no cancer cells detected in the SN on frozen hematoxylin-eosin staining and cytokeratin immunohistochemical staining. We reported this rare case of ductal carcinoma in-situ detected by FDG-PET as past of a medical checkup.     

触诊阴性乳腺病灶术中冰冻诊断的可行性分析

 目的 探讨触诊阴性乳腺病灶活检术中冰冻诊断的准确性与可行性。方法 由钼靶发现的触诊阴性乳腺病灶158例，采用金属线定位技术切除活检，术中进行冰冻切片与诊断，以石蜡组织学诊断为准，评价冰冻诊断的准确性。结果 158例标本中，病理巨检时仅80例（50．6％）发现肉眼可见的异常病灶，平均长径1．2cm。石蜡组织学诊断乳腺浸润癌15例，微小浸润导管癌15例，原位癌12例，导管上皮不典型增生5例，占29．7%（47／158）。术中冰冻对乳腺浸润癌诊断的准确率为93．3％，对微小浸润癌、原位癌、导管上皮不典型增生诊断的准确率分别为60％、58．3％与60％，误诊均为假阴性与低估诊断，无假阳性与过度诊断，原因主要为切片误差与解释错误。结论 冰冻切片对浸润性乳腺癌诊断的准确率高，可用于指导触诊阴性乳腺病灶活检术中手术方案的选择，而对微小浸润癌、原位癌及导管上皮不典型增生常出现假阴性与低估诊断，应待石蜡组织学诊断后再决定手术方案。     

Immunohistochemical analysis on biological markers in ductal carcinomain situ of the breast

BACKGROUND: The increasing use of mammographic screening has led to an increased detection of ductal carcinoma in situ (DCIS) of the breast. The detailed biological characteristics of DCIS and a new classification of DCIS based on these characteristics are needed. METHODS: Immunohistochemical studies were performed to assess the expression of c-erbB-2 (ErbB-2), estrogen receptor (ER), p53 and proliferative activity (Ki-67) in 65 patients with pure DCIS and 60 with invasive ductal carcinoma (IDC). We classified pure DCIS tumors using three classifications, the architectural, Nottingham, and Van Nuys classifications. RESULTS: ErbB-2, ER and p53 staining was positive in 34%, 66% and 21% of patients with DCIS, respectively, and 58%, 42% and 33% in patients with IDC, respectively. Ki-67 stained positively in 1.5% of patients with DCIS and 11.2% of patients with IDC. The comedo type showed a high rate of positive ErbB-2 and p53 staining. The cribriform and papillary types showed a high rate of positive ER staining. Under the Van Nuys classification, ErbB-2, p53 and Ki-67 expression were highest in the group with high nuclear grade and lowest in the group with non-high nuclear grade without necrosis. CONCLUSION: Although the biological markers of IDC tended to suggest aggressive behavior more so than those of DCIS, these differences were based on the histological sub-type, comedo or non-comedo. The Van Nuys classification best defined the subgroups of DCIS with a distinct expression pattern of biological markers, and the best candidates for breast-conserving surgery.     

Three-dimensional mr imaging of mammographically detected suspicious microcalcifications

BACKGROUND: The purpose of this study was to evaluate the diagnostic value and clinical significance of three-dimensional MR imaging of the breast in patients with mammographically detected suspicious microcalcifications. METHODS: Forty patients with suspicious microcalcifications on mammography were evaluated with three-dimensional MR imaging. MR findings were grouped mainly by distribution of abnormal enhancement (linear, focal-clumped, segmental-clumped, segmental-stippled and diffuse-stippled). These findings were compared with the mammography findings according to the criteria of the Breast Imaging Reporting and Data System (BI-RADS) and histopathologic data. RESULTS: Twenty patients had proven malignancies, most frequently ductal carcinoma in situ. For all the cases, linear (100%) and segmental-clumped type (100%) enhancement on MR imaging showed a significantly higher risk for malignancy. Diffuse stippled type (7%) and no enhancement (0%) on MR imaging indicated the lowest possibility of malignancy. 3D-MR imaging showed a sensitivity of 90%, a specificity of 95% and an overall accuracy of 93% in this study. CONCLUSIONS: Three-dimensional MR imaging of the breast can more accurately diagnose ductal carcinoma in situ. Combined with mammography, this procedure is useful for reducing the number of false-positive biopsies and helpful for deciding the better management of patients with mammographically detected suspicious microcalcifications.     

Non-palpable ductal carcinomain situ (DCIS) with microinvasion arising in a radial scar presenting with spiculation alone on mammograms: A case report

A case of ductal carcinoma in situ (DCIS) with microinvasion arising in a radial scar of the breast is presented. A 57-year-old woman visited our hospital with bloody discharge from her left nipple. There were no abnormal findings on cytology, carcinoembryonic antigen (CEA) level of nipple discharge was <500 ng/ml, and mammograms were normal. After 2 years of careful periodic follow-up, spiculation without a central core appeared on mammograms. The CEA level of the nipple discharge increased to 1,000 ng/ml. Ductgraphy showed a connection between the duct with the discharge and the center of the spiculation. Since these findings suggested malignancy, she underwent segmentectomy of the breast, and pathological examination showed a radial scar and DCIS with microinvasion in the ducts within the radiating bands of fibrous tissues. We discuss the characteristics of a radial scar and its relationship to breast cancer based on our experience and a review of the literature.     

Indeterminate calcification and clustered cystic lesions are strongly predictive of the presence of mucocele-like tumor of the breast: a report of six cases

Mucocele-like tumor (MLT) of the breast is a mucinous disorder that is generally difficult to distinguish from mucinous carcinoma. Moreover, MLT is often accompanied by atypical ductal hyperplasia (ADH) or ductal carcinoma in situ (DCIS), and preoperative diagnosis is very confusing. In this paper, we summarize the clinicopathological characteristics of six cases of MLTs, comparing them with mucinous carcinoma. MLTs were characterized by clustered, coarse calcification on mammography and clustered, hypoechoic lesions with or without echogenic spots on ultrasonography. On fine-needle aspiration cytology (FNAC), three cases were correctly diagnosed and three cases were judged to be indeterminate, or insufficient material was present. Ultimately, excisional biopsies were performed to obtain a correct diagnosis in all cases. Immunohistochemical staining of MLTs accompanied by ADH or DCIS (malignant MLTs) revealed the presence of MUC6, while MLTs without ADH or DCIS (benign MLTs) were MUC6-negative. At present, as it is difficult to detect small areas of DCIS adjacent to the MLT by FNAC, excisional biopsy is essential. Immunohistochemical staining for MUC6 may be useful in differentiating between benign MLTs and malignant MLTs, although further investigation is needed.     

Mammary hamartoma

Introduction Mammary hamartomas are rare benign breast lumps. They are usually painless, wellcircumscribed, mobile and with no adherence to skin or muscle, composed of varying amounts of fat, glandular and fibrous tissue. Mammary hamartoma has been classically considered as an underdiagnosed pathology, but with the increasing use of diagnostic procedures in breast tumours, the number of hamartomas has increased in the last years. Because there is no distinct pathological feature, a correlation with the clinical findings and image techniques is necessary in order to achieve a correct diagnosis of the pathology. Materials and methods The clinicopathological features of 8 mammary hamartomas are reported here. Results The patients are ranged in age from 34 to 67 years. The initial manifestation was in all cases a well-circumscribed, soft, palpable breast lump. Mammography was performed in all patients. Other diagnostic procedures used in the diagnosis were Ultrasound, Fine Needle Aspiration Cytology and Needle Core Biopsy. Treatment was tumourectomy. We describe a case of recurrence after excision of the lump in a more aggressive histological form and one patient who presented the coexistence of a mammary hamartoma and an invasive ductal carcinoma. Conclusion Mammary hamartoma is an uncommon breast tumour. It is necessary the correlation between pathology and clinical and radiological findings. We express our management plan for these lesions.     

Evaluation of matrix metalloproteinase-2 (mmp-2) activity with film in situ zymography for improved cytological diagnosis of Breast tumors

BACKGROUND: Fine-needle aspiration (FNA) biopsy of breast tumors is a reliable diagnostic method for identifying breast carcinoma. However, it is sometimes difficult to definitely distinguish malignant from benign lesions. To improve the cytological diagnosis of breast tumors, we investigated the expression of active matrix metalloproteinase-2 (MMP-2), as detected by film in situ zymography (FIZ). METHODS: We evaluated 34 fresh breast tumors, 25 paraffin-embedded breast tissue specimens, and a human cancer cell line (HT1080). MMP-2 expression was determined by immunocytochemistry or immunohistochemistry. Frozen sections and aspiration cytology samples of breast cancer were incubated on gelatin-coated films for the detection of active MMP-2. RESULTS: Immunohistochemistry showed that MMP-2 was expressed in cancer cells and stromal cells, but not in most benign breast lesions. Active MMP-2, but not pro-MMP-2, in the conditioned medium of HT1080 cells showed gelatinolytic activity on FIZ analysis, while the expression of both forms of MMP-2 was detected by immunocytochemistry. Gelatinolytic activity was also detected by FIZ analysis of aspiration cytology samples and frozen sections from the breast cancers, and there was a significant correlation between this gelatinolytic activity and the detection of MMP-2 expression by immunocytochemistry. CONCLUSIONS: The present study demonstrated that measurement of gelatinolytic activity by FIZ analysis of aspiration cytology samples may be useful for improving the cytological diagnosis of breast tumors.     

Differential distribution of ErbB-2 and pS2 proteins in ductal carcinomain situ of the breast

Summary We examined the expression of ErbB-2 and pS2 proteins in 59 ductal carcinomain situ (DCIS) of the breast, either pure DCIS or DCIS associated with invasive carcinoma, using immunohistochemical staining of paraffin-embedded sections. Positive staining for ErbB-2 and pS2 proteins was noted in 32% (19/59) and 46% (27/59) of DCIS, respectively. An inverse relationship between ErbB-2 and pS2 status in DCIS was observed (p < 0.01). From the viewpoint of histological subtype, the prevalence of ErbB-2 protein expression was significantly higher in the comedo subtype than the cribriform-micropapillary subtype. The prevalence of immunoreactivity for ErbB-2 in solid subtype was intermediate between those of the other two groups. In contrast, the prevalence of pS2 expression was significantly lower in the comedo subtype than in the cribriform-micropapillary subtype. Again, the prevalence of pS2 protein expression in the solid subtype was intermediate between those of the other two subtypes. Our results suggest that DCIS is biologically heterogeneous with regard to such marker substances. This has possible implications for management of these lesions.     

Mammotome core biopsy for mammary microcalcification: analysis of 160 biopsies from 142 women with surgical and radiologic followup

BACKGROUND: Although stereotaxic fine-needle aspiration biopsy or core biopsy (14-gauge) have proven to be accurate techniques for the evaluation of mammographically detected microcalcification, the development of the Mammotome Biopsy System (Biopsys Medical, Inc., Irvine, CA) has led many medical centers to use this vacuum-assisted device for the sampling of microcalcification. METHODS: One hundred forty-two women underwent 160 stereotaxic Mammotome core biopsies of mammographic calcification over a 1-year period. The stereotaxic procedure was performed by radiologists using the Mammotome Biopsy System. Microcalcification was evident on specimen radiographs and microscopic slides in 99% of the cases. Excisional biopsy was recommended for diagnoses of atypia or carcinoma. Patients with benign diagnoses underwent mammographic followup. RESULTS: One hundred thirty-two benign, 12 atypical, and 15 adenocarcinoma diagnoses (comprising 1 lobular adenocarcinoma in situ [LCIS], 1 invasive ductal adenocarcinoma [IDC], and 13 intraductal adenocarcinomas [DCIS]: 10 comedo, 1 cribriform, 2 mixed cribriform and micropapillary) were rendered. Surgical excision in eight patients with atypia on Mammotome biopsy (two refused surgery, two were lost to followup) showed ductal hyperplasia in three, atypical ductal hyperplasia (ADH) in three and DCIS (low grade, solid) in two patients. Surgical excisions in 14 patients diagnosed with carcinoma (1 patient lost to followup) showed ADH in 3, ADH and LCIS in 1, residual DCIS in 8, IDC in 1, and microinvasive carcinoma in 1 patient. CONCLUSIONS: A diagnosis of atypia on Mammotome biopsy warranted excision of the atypical area, yet the underestimation rate for the presence of carcinoma remained low. The likelihood of an invasive component at excision was low for microcalcification diagnosed as DCIS on Mammotome biopsy. Mammotome biopsy proved to be an accurate technique for the sampling and diagnosis of mammary microcalcification.     

Immunohistochemical demonstration of metallothionein in normal human breast tissue and benign and malignant breast lesions

Metallothioneins (MTs) are a set of low molecular weight proteins with a high binding affinity to metal ions. MT over-expression has been recently demonstrated in invasive ductal carcinoma of the breast with poor clinical prognosis. In the present study, MTs have been immunohistochemically investigated in normal human breast tissue and a variety of benign, pre-invasive, and malignant breast lesions. In normal breast tissue, MTs were present in myoepithelial cells whereas the vast majority of luminal cells were MT negative. In lesions without increased cancer risk (adenosis and scleradenosis), MT was only immunolocalized in myoepithelial cells. In papillomas, MT was also found exclusively in myoepithelial cells. In most cases of epitheliosis, both the luminal and myoepithelial cells expressed MT. Atypicallobular hyperplasia, lobular carcinomain situ, and 13/15 invasive lobular carcinomas showed no MT over-expression. The two invasive lobular carcinomas with MT over-expression were classified as pleomorphic lobular carcinomas with apocrine differentiation. In contrast to lobular cancerization, 12/24 ductalin situ carcinomas and 9/20 invasive ductal carcinomas showed MT over-expression.In situ components found within invasive ductal carcinomas usually reflected the MT status of their invasive counterpart. It is concluded from our immunohistochemical results that breast carcinoma cases with MT overexpression arise from lesions which also show MT overexpression. Thus MT expression in carcinomas may be regarded as a genuine feature of the tumour cells and seems not to be related to endogenous or exogenous factors known to induce MT synthesis.     

Fluorescentin situ hybridization assessment of chromosome 8 copy number in breast cancer

Summary Conventional cytogenetics of breast and other solid tumors has been hampered by a number of factors. An analysis of breast tumor tissues was therefore undertaken using fluorescentin situ hybridization (FISH). A total of 34 specimens were analyzed using a chromosome 8-specific -satellite probe. Various approaches were tested and compared. Among 30 informative samples, 11 infiltrating ductal carcinomas, not otherwise specified (NOS), 5 ductal carcinomasin situ, 5 lobular carcinomas, 3 papillary carcinomas, and 6 benign lesions were studied. Of the 11 cases of infiltrating ductal carcinomas (NOS) analyzed, four cases showed 3 signals, one case showed 4 signals, and the rest showed 2 signals. Of the 5 cases of ductal carcinomain situ samples, 1 showed 3 signals and the other 4 cases showed 2 signals. All cases of lobular carcinomas, papillary carcinomas, and benign lesions showed 2 signals. We inferred from these data that 36% of the infiltrating ductal carcinomas (NOS) were trisomic and 9% were tetrasomic, whereas 20% of the ductal carcinomasin situ were trisomic. All samples from lobular carcinomas, papillary carcinomas, and the benign lesions were disomic. From our preliminary data, it can further be concluded that a subset of breast cancer is characterized by chromosome 8 trisomy. These data are consistent with an ever-increasing database on the association of chromosomal 8 trisomy with other cancers such as leukemia, lymphoma, prostate cancer, ovarian carcinoma, salivary gland tumor, malignant melanoma, desmoid tumors, and recently gestational trophoblastic disease. It is also noted that the ability to analyze formalin-fixed, paraffin-embedded archival material will enable a more comprehensive cytogenetic study of breast cancer than is currently available.     

Is p63 reliable in detecting microinvasion in ductal carcinoma<Emphasis Type="Italic">in situ</Emphasis> of the breast?

P63, a p53 homologue, is considered to be a marker of myoepithelial cells in breast tissue. This study was carried out to determine the sensitivity of p63 in detecting myoepithelial cells in DCIS and to compare the results obtained with smooth-muscle actin (1A4) in an attempt to verify the reliability of p63 as a possible marker of microinvasion in breast carcinoma. Fifteen DCIS of the breast were submitted to immunohistochemical analysis with anti-p63 and 1A4 antibodies and to a double immunolabeling study using p63 with 1A4. The double immunolabeling study showed that the same cells positive for p63 were also positive for 1A4. The three cases of DCIS with micro-invasion were negative for p63 and 1A4 in the foci of invasiveness. P63 staining was continuous in five of twelve cases of DCIS without microinvasion, being focal and discontinuous in 6 cases and completely negative in one case. Smooth-muscle actin staining was continuous in nine of twelve cases, including the five cases positive for p63. Smooth-muscle actin was focal and discontinuous in only two cases, which were also discontinuous for p63. The DCIS negative for p63 was also negative for 1A4. In conclusion, our results confirm the data of literature that p63 is a specific marker of myoepithelial cells in breast tissue. However, p63 is not as sensitive as 1A4 in staining myoepithelial cells and lack of p63 expression cannot be used as a reliable marker of invasiveness in ductal carcinoma in situ of the breast.