Immunohistochemical evaluation of c-erbB-2 oncogene expression in ductal carcinoma in situ and atypical ductal hyperplasia of the breast

Cases of ductal carcinoma in situ (DCIS) and atypical ductal hyperplasia (ADH) of the breast were examined for expression of the protein product of the c-erbB-2 (neu, HER-2) oncogene using two different polyclonal antibodies via an avidin-biotin immunoperoxidase method on formalin- or Bouin'-fixed, paraffin-embedded tissue. Fifty-five percent (18/33) of DCIS and 10% (2/21) of ADH were positive. Significant c-erbB-2 expression in DCIS was generally divided on histologic grounds: ten of ten comedocarcinomas showed strong membrane staining, while only one of 14 small cell DCIS cases (micropapillary or cribiform patterns) showed immunostaining (which was weak and basilar in this single case). DCIS cases of mixed histology were strongly positive in areas of comedocarcinoma. In two of three cases of associated Paget's disease strong membrane staining was seen. The two c-erbB-2-positive ADH cases showed weak basilar staining akin to the small cell DCIS cases. Five cases of lobular neoplasia (atypical lobular hyperplasia or lobular carcinoma in situ) associated with DCIS or ADH were negative for c-erbB-2 expression. We conclude that comedocarcinoma in situ and Paget's disease frequently express the c-erbB-2 protein and are both histologically and biochemically distinct from ADH and small cell patterns of DCIS. We advocate precise subclassification of DCIS on histopathologic reports, particularly in view of reports that overexpression of the c-erbB-2 oncogene in infiltrating breast carcinomas may be associated with a poor prognosis.     

c-erbB-2 expression in different histological types of invasive breast carcinoma.

Sections of 149 breast carcinomas were examined for the over-expression of c-erbB-2 oncoprotein using the avidin-biotin immunoperoxidase technique and two different specific antibodies. These included the polyclonal antibody 21N and the monoclonal antibody 4D5. The tumours were divided into two main groups. The first included 75 cases of invasive ductal and classic invasive lobular carcinomas. The second group consisted of 74 cases with histological types known to have a good prognosis, including mucinous, alveolar variant of invasive lobular, medullary, tubular, cribriform and papillary carcinomas. Fifteen (20%) tumours of the first group were positive with the two antibodies. Fourteen of these were of the ductal type and one was a mixed invasive ductal and lobular carcinoma. Ten of the pure ductal cases had areas of comedo carcinoma. The intraductal elements in a further tumour were positively stained with 21N antibody only. None of the second group of tumours, which included histological types known to have good prognosis, stained with 4D5, although one mucinous carcinoma was positively stained with 21N. These findings suggest that in invasive breast carcinoma immunostaining for c-erbB-2 is mainly seen in a subgroup of ductal tumours, and that almost all other histological types, especially those associated with good prognosis, lack this expression.     

Epidermal growth factor receptor immunohistochemistry in different histological types of infiltrating breast carcinoma.

AIMS--To determine the immunohistochemical expression of epidermal growth factor receptor (EGF-R) in high grade, intermediate, and low grade tumours. METHODS--Specimens from 931 breast carcinomas were partly formalin fixed and paraffin wax embedded, to classify cases, and partly frozen in liquid nitrogen, cryostat sectioned, and immunostained using two monoclonal antibodies from clone 455 and 528 to demonstrate EGF-R positive cells. An avidin-biotin complex and peroxidase method was used after incubation with biotinylated anti-mouse antibody; colour was developed using a diaiminobenzidine solution. RESULTS--Low grade carcinomas seldom expressed EGF-R (n = 3) compared with 106 high grade infiltrating ductal carcinomas: EGF-R positive cases were much less common in infiltrating lobular than in infiltrating ductal carcinoma. Medullary carcinomas did not differ from infiltrating ductal carcinomas. CONCLUSIONS--The very low incidence of EGF-R positive cases in the "special type" group of breast carcinomas with a good prognosis is in line with the absence of the homologous c-erbB-2 and p53 oncoproteins, and the rarity of highly proliferating and oestrogen/progesterone negative cases. EGF-R expression in infiltrating lobular carcinoma was in keeping with the intermediate behaviour of this kind of tumour. EGF-R expression in cases of pure medullary carcinoma is the same as that of high grade tumours.     

The c-erbB-2 Protein in Primary and Metastatic Breast Carcinomas

Material from 41 patients with primary breast carcinoma and lymph node metastases at the time of primary surgical intervention was immunostained for c-erbB-2 protein, neuron-specific enolase (NSE), and estrogen receptors. Thirty of the primary breast carcinomas were of ductal type. Six were classified as infiltrating lobular carcinomas, 2 were apocrine, 1 was mucinous, and 1 was a tubular carcinoma. One tumor could not be classified as ductal or lobular by light microscopic examination alone. The number of lymph node metastases available varied from 1 to 14 per case (median, 3.9). Nine (22%) of the primary breast carcinomas (8 ductal and 1 apocrine) expressed c-erbB-2 protein and showed c-erbB-2 gene amplification; 12 expressed NSE immunoreactivity. None expressed both markers. Estrogen receptor immunoreactivity was present in 23 of the 41 cases, including 9 of the NSE-positive cases. C-erbB-2 protein-positive metastases were present in 18 cases (44%), and in 13 cases all metastases were immunostained. In 5 cases the expression of c-erbB-2 protein varied from metastasis to metastasis. NSE immunoreactivity was expressed in 10 cases, and in 3 cases with minor NSE-positive cell populations the metastatic lesions expressed c-erbB-2 protein as well. All 9 primary breast carcinomas expressing c-erbB-2 protein had lymph node metastases with c-erbB-2-immunoreactive tumor cells. Eight of the 9 c-erbB-2 protein-negative primary tumors with metastases expressing c-erbB-2 protein showed no amplification of the c-erbB-2 gene. Thus expression of c-erbB-2 protein can occur during the metastatic process, even if it seems to be missing in the primary tumor. On the other hand, if a primary breast carcinoma expresses c-erbB-2 protein, this feature seems to be present in all the tumor metastases as well.     

The Quarterly Case: Metastatic Tumor of Unknown Origin

Material from 41 patients with primary breast carcinoma and lymph node metastases at the time of primary surgical intervention was immunostained for c-erbB-2 protein, neuron-specific enolase (NSE), and estrogen receptors. Thirty of the primary breast carcinomas were of ductal type. Six were classified as infiltrating lobular carcinomas, 2 were apocrine, 1 was mucinous, and 1 was a tubular carcinoma. One tumor could not be classified as ductal or lobular by light microscopic examination alone. The number of lymph node metastases available varied from 1 to 14 per case (median, 3.9). Nine (22%) of the primary breast carcinomas (8 ductal and 1 apocrine) expressed c-erbB-2 protein and showed c-erbB-2 gene amplification; 12 expressed NSE immunoreactivity. None expressed both markers. Estrogen receptor immunoreactivity was present in 23 of the 41 cases, including 9 of the NSE-positive cases. C-erbB-2 protein-positive metastases were present in 18 cases (44%), and in 13 cases all metastases were immunostained. In 5 cases the expression of c-erbB-2 protein varied from metastasis to metastasis. NSE immunoreactivity was expressed in 10 cases, and in 3 cases with minor NSE-positive cell populations the metastatic lesions expressed c-erbB-2 protein as well. All 9 primary breast carcinomas expressing c-erbB-2 protein had lymph node metastases with c-erbB-2-immunoreactive tumor cells. Eight of the 9 c-erbB-2 protein-negative primary tumors with metastases expressing c-erbB-2 protein showed no amplification of the c-erbB-2 gene. Thus expression of c-erbB-2 protein can occur during the metastatic process, even if it seems to be missing in the primary tumor. On the other hand, if a primary breast carcinoma expresses c-erbB-2 protein, this feature seems to be present in all the tumor metastases as well.     

Assessment of invasion in breast lesions using antibodies to basement membrane components and myoepithelial cells.

This paper describes immunostaining of consecutive sections from 15 cases of fibrocystic change of the breast (including 2 examples of intraductal papilloma), 4 ductal carcinomas-in-situ and 17 invasive carcinomas (4 tubular, 1 papillary, 2 lobular and 10 infiltrating ductal, NOS) with antisera to components of the basement membrane (BM), type IV collagen and laminin, and with the muscle antibodies actin and muscle-specific actin. A simple digestion technique was developed to improve the clarity of BM staining with these antibodies. The BM stains facilitated identification of small invasive foci through breaks in the BM in 2 of the cases which had been reported as pure intraductal carcinoma. Tubular carcinomas were surrounded by abnormal, fragmented, and focally discontinuous BM, a feature which could be used to distinguish this well-differentiated breast carcinoma sub-type from sclerosing adenosis, in which individual acini were invariably surrounded by a continuous BM. BM staining emphasized the fibrovascular core of intraductal papillomas, whereas the BM layer was absent in intraductal, cytologically malignant, papillary projections. Similarly, myoepithelial cells, stained with antisera to muscle actins, were identified in a continuous layer surrounding benign epithelial proliferations. These immunohistochemical staining techniques may thus assist the diagnostic histopathologist in differentiating between benign epithelial proliferations of the breast and well-differentiated invasive breast carcinoma, and in identifying foci of microinvasive carcinoma.     

Detection of amplified int-2/FGF-3 gene in primary breast carcinomas using differential polymerase chain reaction.

Amplification of int-2/FGF-3 gene was investigated by differential polymerase chain reaction (dPCR) in 440 archival primary breast carcinoma tissues. Of these, 23 were comedo ductal carcinoma in situ (DCIS), 18 were non-comedo DCIS, 41 were comedo DCIS with adjacent invasive ductal carcinomas, 19 were non-comedo DCIS with adjacent invasive ductal carcinomas, 270 were invasive ductal carcinomas, 33 were invasive lobular carcinomas, 21 were colloid carcinomas and 15 were medullary carcinomas. Int-2 was amplified in 22% (96/440) of the primary breast carcinomas. It was shown that int-2 was amplified in 13% (3/23) of the comedo DCIS, 17% (7/41) of the comedo DCIS and 29% (12/41) of the adjacent invasive ductal carcinomas, 26% (71/270) of the invasive ductal carcinomas, 18% (6/33) of the invasive lobular carcinomas, 10% (2/21) of the colloid carcinomas and 13% (2/15) of the medullary carcinomas. In contrast, int-2 was not amplified in non-comedo DCIS and invasive ductal carcinomas with adjacent non-comedo DCIS lesions. A significant association was observed between int-2 amplification in the in situ components and adjacent invasive lesion (P<0.05). All tumors with int-2 amplification in the in situ lesions (7/7) also demonstrated same degree of amplification in the adjacent invasive components. However, 9% (5/53) of the tumors with no amplified int-2 gene in the in situ components showed int-2 amplification in the adjacent invasive lesions. A significant relationship was noted between amplification of int-2 and lymph node metastases (P<0.05) and poorly differentiated tumors (P<0.05) but not with estrogen receptor status (P>0.05) and proliferation index (Ki-67 and PCNA) (P>0.05). In Malaysia, majority of the patients belong to younger age group (<50 years old) but a comparison of the age groups showed that the amplification of int-2 was not statistically associated with patient age (P>0.05). These observations indicate that amplification of int-2 tends to strengthen the view that int-2 may have the potential to be an indicator of poor prognosis regardless of the age of the patient. Moreover, the presence of int-2 amplification in preinvasive, preinvasive and adjacent invasive lesions, and invasive carcinomas suggest that int-2 could be a marker of genetic instability occurring in early and late stages of tumor development.     

Breast carcinoma in women 35 years and younger: a pathological study

AIMS: To document the pathological features of breast carcinoma diagnosed in women aged 35 years or less. METHODS: The files of the Department of Pathology, Singapore General Hospital, were searched for cases of breast cancer diagnosed in individuals aged 35 years or less between January 1993 and December 2004. Histological slides and pathology reports were retrieved and reviewed. Pathologic parameters of tumour size, histological grade, accompanying ductal carcinoma in situ (DCIS), lymphovascular invasion, nodal status, hormone receptor and c-erbB-2 profiles, were documented. RESULTS: Of 112 cases of breast cancer, 91 (81.3%) were invasive carcinomas, 17 (15.2%) pure DCIS, three (2.7%) were diagnosed on needle aspirates. No residual tumour was available for microscopic assessment in one (0.9%) patient who underwent bilateral mastectomy in our institution but had her initial surgery elsewhere. Invasive tumour size ranged from 0.3 to 11.5 cm (mean 2.7 cm, median 2.1 cm), with 84 (92.3%) infiltrative ductal, two (2.2%) lobular, two (2.2%) mucinous, two (2.2%) atypical medullary, and one (1.1%) mixed ductal-lobular. The majority were grade 3 (54 cases, 59.3%), with 24 (26.4%) grade 2 and 7 (7.7%) grade 1, while grading was not accomplished in six (6.6%) cases. Nodal status was positive in 39 (42.9%), negative in 25 (27.4%), unknown in the rest (27 cases, 29.7%). Oestrogen and progesterone receptors (ER, PR) were positive in 51 (61.4%) and 43 (51.8%) cases, respectively, out of 83 (91.2%) cases in which they were evaluated. c-erbB-2 immunostaining, carried out in 54 (59.3%) invasive cancers, showed positivity in 16 (29.6%) cases.DCIS cases ranged from 0.25 to 6.2 cm (mean 2.2 cm, median 2 cm) in size. Nuclear grade was low in seven (41.2%), intermediate in four (23.5%), and high in six (35.3%). CONCLUSION: The majority of breast carcinomas in young women are invasive, with T2 disease at presentation, and of poor histological grade. The recent rise in numbers suggests increased detection, plausibly due to improved awareness of breast disease among the younger female population. Pathogenetic causes that differ from breast carcinogenesis in older women have to be further clarified.     

Expression of EGFR Family and Steroid Hormone Receptors in Ductal Carcinoma in situ of the Breast

The expression of EGFR family and steroid hormone receptors was examined in a series of 40 cases of pure ductal carcinoma in situ (DCIS) of the breast by immunohistochemical staining of paraffin-embedded sections. Hematoxylin and eosin-stained sections were used to classify the tumors according to the published criteria by Holland et al. (Holland R, Peterse JL, Millis RR, et al. Semin Diagn Pathol. 1994;11:167-180). Of the tumors 48% were immunoreactive for EGFR, 63% for c-erbB-2, 78% for c-erbB-3, 95% for c-erbB-4, 88%for estrogen receptor (ER) and 80% for progesterone receptor (PR). Statistically significant association between histological grade (differentiation) and c-erbB-2 protein expression was seen ( p <.001). In addition, expression of c-erbB-4 protein was associated with c-erbB-2 ( p =.004), c-erbB-3 ( p =.058), ER ( p =.002) and PR ( p =.004). It is concluded that c-erbB-2 expression in DCIS is associated with high-grade pathological features, and a higher c-erbB-2 expression is seen in DCIS than in invasive breast carcinomas. A possible association between extensive expression of c-erbB-4 and steroid hormone receptors in proliferative and premalignant breast epithelial cells and the c-erbB-2 expression in DCIS and invasive breast carcinomas is discussed.     

Expression of EGFR family and steroid hormone receptors in ductal carcinoma in situ of the breast.

The expression of EGFR family and steroid hormone receptors was examined in a series of 40 cases of pure ductal carcinoma in situ (DCIS) of the breast by immunohistochemical staining of paraffin-embedded sections. Hematoxylin and eosin-stained sections were used to classify the tumors according to the published criteria by Holland et al. (Holland R, Peterse JL, Millis RR, et al. Semin Diagn Pathol. 1994;1 1:167-180). Of the tumors 48% were immunoreactive for EGFR, 63% for c-erbB-2, 78% for c-erbB-3, 95% for c-erbB-4, 88% for estrogen receptor (ER) and 80% for progesterone receptor (PR). Statistically significant association between histological grade (differentiation) and c-erbB-2 protein expression was seen (p <.001). In addition, expression of c-erbB-4 protein was associated with c-erbB-2 (p=.004), c-erbB-3 (p=.058), ER (p=.002) and PR (p=.004). It is concluded that c-erbB-2 expression in DCIS is associated with high-grade pathological features, and a higher c-erbB-2 expression is seen in DCIS than in invasive breast carcinomas. A possible association between extensive expression of c-erbB-4 and steroid hormone receptors in proliferative and premalignant breast epithelial cells and the c-erbB-2 expression in DCIS and invasive breast carcinomas is discussed.     

Mucins in breast carcinoma

Mucins produced by breast carcinoma have been shown to be neutral and sialomucins. Colonic mucins were found to be sulfated with an o-acylated sialomucin component that was reported by Culling to be specific to normal colonic epithelium and colonic adenocarcinoma. This type of molecule gives a positive staining after potassium hydroxide/periodate borohydride (KOH/PBT) treatment. We studied the types of mucin in five medullary carcinomas, 10 infiltrating duct carcinomas, and 10 infiltrating lobular carcinomas of the breast. Twenty-two of 25 cases were positive for neutral mucins; 14 of 25 were positive for sialomucin. Sulfated mucins were found in five of 10 cases of infiltrating lobular carcinoma, and in six of 10 cases there was a positive periodic acid-Schiff reaction following KOH/PBT. This study illustrates the presence of sulfated mucin and o-acylated sialomucin in infiltrating lobular carcinoma of the breast. Thus, mucin stains cannot be used in differentiating metastatic breast carcinoma from colonic carcinoma. The significance of this finding and its relation to tumor histogenesis, typing, and biologic behaviors require further analysis.     

Tubular carcinoma of the breast and associated intra-epithelial lesions: a comparative study with invasive low-grade ductal carcinomas

Ductal intra-epithelial lesions of the breast are associated with invasive neoplasms and comprise a large spectrum of histological patterns. We have examined 23 cases of pure tubular carcinomas (TCs) of the breast and 53 cases of invasive ductal low-grade carcinomas to determine the relationship and distribution of intra-epithelial lesions, mainly of ductal in situ carcinoma type, but including also lobular intra-epithelial neoplasia (LIN) in both entities. Eleven cases of TC showed flat epithelial atypia (FEA) (47.8%), and, in 14 and 6 cases, micropapillary and cribriform low-grade ductal carcinoma in situ (DCIS) were present (60.7 and 26.1%, respectively). On the opposite, in ductal grade I invasive carcinomas, the most frequent architectural pattern was low-grade DCIS growing in arcades in 26 cases (49%). While absent in TCs, low-grade DCIS of solid type was found in five (9.4%) cases of ductal invasive carcinomas, where FEA were present in seven (13.2%) cases. LIN lesions were present in four (17.4%) cases of TC, whereas they represented 7.5%, as reported by Carstens et al. (Am J Clin Pathol 58:231–238, 1972), of cases of low-grade carcinomas. These results suggest that invasive pure TC and low-grade ductal carcinomas of the breast are different lesions, and support the fact that TC, of low histopathological grade, is a particular distinct tumoural entity.     

Incidental finding of mammary carcinoma in lumpectomy specimens

More and more breast lumpectomies are being performed due to mammographic screening, and both in situ and invasive breast carcinomas are being detected earlier and smaller in size. The objective of this study was to determine the presence of incidental microscopic breast carcinoma in mammography-guided lumpectomy specimens. A prospective study was carried out by processing in surgical pathology approximately 9,000 breast lumpectomy specimens during a 2.5-yr period so that the entire specimens were embedded for microscopic examination. Excluded from the study were cases with grossly or microscopically identified carcinomas greater than 10 mm2, and non-invasive carcinomas diagnosed in association with invasive carcinoma. Cases with multifocal carcinomas, prior diagnosis of breast cancer, or prior history of breast biopsy were also excluded. Carcinomas present in the same tissue blocks as the clinically suspected lesions such as palpable nodules, microcalcification, or other mammographic abnormalities were excluded as well. Fifty cases of incidental microscopic mammary carcinoma were found including 8 infiltrating ductal carcinomas (IDC), 2 infiltrating lobular carcinomas (ILC), 21 intraductal carcinomas (DCIS), and 19 lobular carcinomas in situ (LCIS). All of the lesions were solitary, located in indistinct loosely arranged fibrous and adipose stromal tissues, and the majority of them were near or at the inked excisional margins. Physicians who care for patients with breast cancer should be aware of the existence of these minute breast carcinomas that are often near or at the surgical margins. The significance of these microscopic findings for therapeutic strategy and prognosis should be determined by long-term follow-up.     

Breast calcifications with percutaneous vacuum-assisted biopsy diagnosis of malignancy or atypical hyerplasia: correlations with surgical findings

Percutaneous, stereotactic, vacuum-assisted biopsy has become a widely used alternative to open surgical biopsy for the initial diagnosis of breast calcifications. We retrospectively assessed the accuracy of the technique in the diagnoses of malignancy and atypical hyperplasia by correlation with the findings of the subsequent surgical excision. We studied 330 consecutive cases of breast calcifications, 216 (65.5%) of which were determined to be benign and 114 (34.5%) to be malignant or atypical at vacuum-assisted biopsy using an 11 gauge instrument. Of the latter 93 were available for comparison with the subsequent surgery, the specific diagnoses as revealed by percutaneous biopsy were as follows: 11 cases of atypical ductal hyperplasia (ADN), 67 cases of ductal carcinoma in situ (DCIS), 6 infiltrating ductal carcinomas (IFDC), 2 cases of atypical lobular hyperplasia and 7 of lobular carcinoma in situ (LCIS). At histological analysis after surgical excision, 3 (27%) of 11 cases previously diagnosed as ADH and 6 (9%) of 67 cases diagnosed as DCIS were shown to actually be higher grade lesions (DCIS/IFDC and IFDC, respectively). Of the 7 lesions diagnosed at vacuum-assisted biopsy as LCIS, surgery and histological analysis showed one infiltrating globular carcinoma and two DCIS. A total of 21 lesions (4 ADH, 14 DCIS, 1 IFDC, 2 LCIS) were completely removed at percutaneous biopsy; the remaining cases were found totally concordant. These data Indicate a substantial accuracy of the percutaneous biopsy: some lesions (particularly those thought to be ADH and DCIS) can be underestimated for sampling error.     

Lobular neoplasia in breast core needle biopsy specimens is not associated with an increased risk of ductal carcinoma in situ or invasive carcinoma

Recent reports suggest that the finding of lobular neoplasia (atypical lobular hyperplasia [ALH] or bular carcinoma in situ [LCIS]) in breast core needle biopsy specimens may be associated with an increased risk of both ductal carcinoma in situ (DCIS) or invasive carcinoma at excision. We reviewed our breast core biopsy material to see if we could confirm this finding. from 4,297 biopsies, 71 cases of lobular neoplasia lone and 35 cases of lobular neoplasia associated with typical ductal hyperplasia were identified. Biopsy follow-up revealed DCIS or invasive carcinoma in none of 6 cases of ALH, none of 9 cases of LCIS, and DCIS in 1 of 11 cases with both atypical ductal hyperplasia and LCIS. Our results suggest that patients with lobular eoplasia in breast core biopsy specimens are not at increased risk of either DCIS or invasive carcinoma at excision, and patients with this finding and no other linical or pathologic indications for biopsy can be llowed up rather than routinely undergo excision.     

Vimentin is preferentially expressed in high-grade ductal and medullary, but not in lobular breast carcinomas.

Two hundred sixty-two invasive breast carcinomas dating from 1979 to 1984 were tested for vimentin and keratin on formaldehyde-fixed paraffin-embedded sections. None of 26 lobular carcinomas expressed vimentin. Vimentin expression in 10% or more of tumor cells was found in 78% of medullary (14 of 18), in 16% of ductal not otherwise specified (NOS) (35 of 214), and in two of four mucinous carcinomas. A further seven tumors showed vimentin expression in less than 1% to 10% of the cells. Vimentin was expressed in tumor cells of 30% (28 of 93) of grade III invasive ductal NOS carcinomas versus 7% (7 of 105) of grade II and 0% of grade I carcinomas (0 of 10). Vimentin was found to be preferentially expressed in tumors growing in broad, often anastomosing bands or sheets with extensive necrosis, scanty supportive stroma, high nuclear grade, and numerous mitoses. The authors conclude that vimentin is not detected in lobular carcinomas, but is preferentially expressed in medullary and in high-grade ductal NOS breast carcinomas.     

Correlation of E-cadherin expression with differentiation grade and histological type in breast carcinoma.

Recently, a correlation has been suggested between a loss of E-cadherin (E-CD) and increased invasiveness of neoplastic cells. In this study, E-CD expression in breast cancer was investigated using an affinity-purified antibody (ECCD-2) in an immunoenzymatic (avidin-biotin-alkaline phosphatase) test. Intensity and extension of E-CD immunoreactivity were evaluated in 61 breast carcinomas and correlated with their histological type and grade, nodal involvement, and hormonal receptor status. Histological types were infiltrating ductal carcinoma of no special type (n = 54) and infiltrating lobular carcinoma (n = 7). All infiltrating ductal carcinomas of no special type except two grade 3 carcinomas showed positive immunoreactivity that was variable among different cases. Grade 1 breast carcinomas (n = 10) showed greater immunoreactivity than grade 2 (n = 25) and grade 3 (n = 19) carcinomas. E-CD immunoreactivity correlated positively with the degree of tubular formation and inversely with the mitoses number. None of the infiltrating lobular carcinomas expressed E-CD in their infiltrating cells, whereas they showed only weak immunostains in areas of atypical lobular hyperplasia and lobular carcinoma in situ. These results indicate that E-CD expression correlates with histological type and grade in breast carcinomas.     

Immunohistochemical distribution of c-erbB-2 in in situ breast carcinoma--a detailed morphological analysis

An immunohistochemical study of c-erbB-2 expression was carried out on in situ (non-invasive) breast carcinoma, using antibody 21N, raised to the intracytoplasmic domain of the c-erbB-2 oncogene product. Strong membrane staining was observed in 44 out of 74 (59 per cent) cases of ductal carcinoma in situ (DCIS), but none of 48 lobular carcinoma in situ (LCIS) lesions. A detailed comparative morphological evaluation using several different parameters, including histological subtypes, was performed within the DCIS group. The results showed that there was a significant correlation between c-erbB-2 expression and the presence of large cell size, periductal lymphoid cell infiltration, marked nuclear pleomorphism, multinucleation, and a high mitotic rate. Of these, cell size appears to be the most important predictor of c-erbB-2 status, followed by the presence of periductal lymphoid cell infiltration. These results indicate, firstly, that LCIS and DCIS are biologically (as well as histologically) different and, secondly, that a subgroup of DCIS, which is associated with c-erbB-2 over-expression, exists and appears to have distinct histological features. The subgroup of DCIS cases which over-express c-erbB-2 may be a biologically definable category with prognostic importance. These results may therefore have relevance to breast screening programmes, but a larger study incorporating clinical data would be necessary to correlate these findings with clinical outcome.     

Frequent E-cadherin gene inactivation by loss of heterozygosity in pleomorphic lobular carcinoma of the breast.

Pleomorphic lobular carcinoma of the breast is a variant of infiltrating lobular carcinoma that has poor prognosis. The pleomorphic appearance of this variant hinders its correct identification and differentiation from ductal carcinoma. The analysis of E-cadherin glycoprotein expression is a powerful tool for distinguishing lobular from ductal carcinomas, because complete loss of E-cadherin expression occurs in most infiltrating lobular tumors and lobular carcinomas in situ, but not in ductal tumors. In the present study, we have evaluated E-cadherin expression by immunohistochemistry in a series of 29 pleomorphic lobular breast carcinomas, including 7 cases with an in situ component. Complete loss of E-cadherin expression was observed in all the cases (29/29, 100%), in invasive and in situ components. To understand better the mechanisms underlying E-cadherin inactivation in this tumor type, the frequency of loss of heterozygosity at the E-cadherin gene locus (16q22.1) was analyzed. All informative tumors (27/27, 100%) showed loss of heterozygosity, thus implying a strong association between loss of E-cadherin expression and loss of heterozygosity at 16q22.1. Moreover, loss of heterozygosity was detected in all in situ components analyzed. These results imply that in terms of E-cadherin inactivation, pleomorphic lobular tumors are identical to classic infiltrating lobular carcinomas and distinct from ductal tumors, and therefore they should be considered a variant of lobular carcinoma of the breast, despite their aggressive behavior.