Pseudoangiomatous hyperplasia of male breast.

Ninety-three male breast specimens have been examined for the presence of pseudoangiomatous hyperplasia of the mammary stroma which has hitherto been described almost exclusively in females. Forty-four cases (47.4%) showed some degree of hyperplasia, varying from small microscopic foci to extensive change involving 90% of the mammary tissue. All but one were found in association with gynaecomastia, early and intermediate stage. The association between pseudoangiomatous hyperplasia and benign proliferative lesions mirrors that reported in the female breast, and our findings suggest that the change may represent a stage in the maturation of newly formed mammary stroma.

Potential use of monoclonal antibodies in the diagnostic distinction of gynecomastia from breast carcinoma in men.

Immunohistochemical (IHC) assays using the monoclonal antibodies (MoAbs) B72.3 and B6.2, recognizing two distinct and independently expressed breast tumor-associated antigens (BTAAs), recently have been shown to significantly improve the accuracy of cytodiagnosis of breast nodules by fine-needle aspiration (FNA). To evaluate whether the same method may be useful diagnostically in distinguishing gynecomastia from breast cancer in men, a retrospective avidin-biotin immunoperoxidase assay study was performed on 50 cases of gynecomastia and 30 cases of breast carcinoma in men, using a panel of five MoAbs known to recognize different BTAAs. The results of this study demonstrated that MoAbs B1.1, HMFG2, and MBr1 displayed a strong reactivity with gynecomastia and carcinoma, but MoAbs B72.3 and B6.2 separated benign and malignant lesions in a high percentage of cases. When used in combination, the latter two reagents reacted with 96% of the carcinomas that were analyzed but labeled only 67% of gynecomastia cases. Thus, the conjoint use of these two reagents may enhance the use of FNA biopsy as a valuable tool in the presurgical diagnosis of breast nodules in men.

Glial fibrillary acidic protein immunoreactivity in normal and diseased human breast.

Immunostaining for glial fibrillary acidic protein (GFAP) identifies a minor subpopulation of immunoreactive myoepithelial cells in the normal resting human breast. The GFAP-immunoreactive cells also express a panel of myoepithelial cell markers, including cytokeratin 14 (CK 14), vimentin, smooth-muscle-specific actin isoforms, nerve growth factor receptor (NGFR) and common acute lymphoblastic leukaemia antigen (CALLA). The percentage of GFAP-immunoreactive myoepithelial cells is greatly increased in various neoplastic and non-neoplastic diseases of the breast, being highest in adenomyoepitheliomas. Furthermore, in all the instances of fibroadenoma, phyllodes tumour, epitheliosis and gynaecomastia, a variable number of epithelial cells also acquires immunoreactivity for GFAP, vimentin, CK 14, NGFR and, to a lesser extent, for CALLA. Conversely, GFAP immunoreactivity has never been encountered in the malignant cells of the different types of breast carcinoma. These findings suggest that the expression of GFAP might be a (possibly transient) feature of proliferating epithelial and myoepithelial cells in breast diseases other than carcinomas.

Variability in surface antigen expression of human breast epithelial cells cultured from normal breast, normal tissue peripheral to breast carcinomas, and breast carcinomas.

Single-cell heterogeneity and variability in expression of several surface antigens on human mammary epithelial cells in short-term culture were studied with immunofluorescence techniques, using polyclonal and monoclonal antibodies. The cultures, derived from normal breast, a fibroadenoma, a gynecomastia, normal breast tissue peripheral to breast carcinomas, and breast carcinomas and their metastases, were studied after one passage in vitro. The percentage of positive cells varied considerably from one tissue sample to another in all categories from normal to malignant, although there was an overall trend toward a decreasing percentage of positive cells of malignant tissues. The relative antigen content varied 3- to 8-fold among individual samples of cells from normal, peripheral, and carcinoma tissue, while the mean values in the three categories were similar. The single-cell variability in relative antigen content was considerable in all individual samples of normal, peripheral, and carcinoma tissues, as reflected in the high coefficients of variation. However, the coefficients of variation were significantly higher for cells from carcinoma and peripheral tissues [69 +/- 10% (S.E.) and 75 +/- 9%, respectively] than for cells from normal breast (48 +/- 5%). By analyzing in clonal colonies the appearance of quantitative variants in expression of a specific surface antigen, detected with a monoclonal antibody, the carcinoma cells were found to have a 10-fold higher rate of phenotypic variability (mean, 1.21 X 10(-2)/cell/generation) than did cells from normal breast (mean, 0.119 X 10(-2)) and one gynecomastia (0.045 X 10(-2)). Mammary epithelial cells from apparently "normal" tissue peripheral to a carcinoma had an intermediate rate of phenotypic variability (mean, 0.310 X 10(-2)) that was significantly higher than that of the normal tissue.

A human breast tissue-associated antigen detected by a monoclonal antibody.

Two monoclonal antibodies were produced in mice immunized with the human breast carcinoma cell line MCF-7. One antibody (24-17.1) reacted with MCF-7 and other breast tumor cell lines and detected an antigen of Mr 95,000. This antigen was not breast-specific because other tumor cell lines were also reactive. The second antibody (24-17.2) detected an antigen of Mr 100,000 (initially appearing to be specific for breast tissue and possibly for breast carcinomas) which was present on 10 of 10 malignant breast lines and absent from 41 of 43 other cell lines of differing origins. The antigen could not be detected by absorption or a direct test on normal tissues (liver, kidney, heart, spleen) or on lymphocytes. In addition, the 24-17.2 antibody reacted absorptively with 12 of 13 fresh breast carcinoma samples but not with fresh colon carcinoma samples. The 100,000-Mr antigen detected by the 24-17.2 antibody appeared to be distinct from the other components of normal breast, such as casein, lactalbumin, or milk fat globulin protein. This evidence indicated that the 24-17.2 antibody detected a human breast carcinoma-associated antigen (HBCAA). However, further histologic studies were used to determine the cellular distribution of the HBCAA, which was found on malignant breast epithelium, the epithelium of gynecomastia, and in lesser amounts and differently distributed on normal breast epithelium. The antigen was also found in several other tissues; nonetheless, the anti-HBCAA could be detected in increased amounts in the sera of patients with breast cancer.

Immunohistochemistry of a gross cystic disease fluid protein (GCDFP-15) of the breast. A marker of apocrine epithelium and breast carcinomas with apocrine features.

Gross cystic disease fluid is a pathologic secretion from breast composed of several glycoproteins, including a unique 15,000-dalton monomer protein, GCDFP-15. By the immunoperoxidase technique, GCDFP-15 was localized in the apocrine metaplastic epithelium lining breast cysts and in apocrine glands in the axilla, vulva, eyelid, and ear canal. In normal breast tissue, a few individual epithelial cells within lobules and small ducts were focally positive for GCDFP-15. Fourteen of 30 breast carcinomas stained positively for GCDFP-15. Of 16 carcinomas with apocrine features, 12 stained positively. Benign and malignant lesions from other tissues, including lung, colon, ovary, endometrium, stomach, prostate, liver, esophagus, and kidney, revealed no immunoreactivity. The only cells of "non-apocrine" tissues that contained GCDFP-15 were serous cells of the submandibular salivary gland, submucosal glands of the bronchi, and accessory lacrimal glands. Phylogenetically, these tissues have biologic features in common with apocrine glands. This report is the first to characterize GCDFP-15 as a specific tissue marker of apocrine epithelium.

Monoclonal antibodies to the human mammary gland. I. Distribution of determinants in non-neoplastic mammary and extra mammary tissues.

Mouse monoclonal antibodies have been raised to the human milk fat globule membrane. The distribution of the antigens detected by four of the antibodies has been examined in formalin-fixed paraffin-embedded human tissues by light microscopic immunocytochemistry. The four antibodies stain lactating breast and normal resting breast. Two exclusively stain the luminal membranes of breast epithelial cells. A third antibody stains in addition the lateral membranes of duct epithelial cells. The fourth antibody stains both epithelial and myoepithelial cells. None of the antibodies is breast specific, nor do they stain every epithelial cell within the breast. Instead, each antibody reveals a complex and heterogeneous distribution of staining throughout the normal tissues. Within the breast, the staining by a given antibody is usually segmental and conforms to secretory units and their associated ducts. Similarly heterogeneous patterns of staining are also observed in the extramammary normal tissues. Despite the apparent morphological identity between breast epithelial cells when examined by conventional light microscopy, the hitherto unrecognised "functional" heterogeneity, which has been revealed by the monoclonal antibodies could have importance in understanding the biology of the normal breast and the pathology of breast cancer.

Isoantigens A, B and H in benign and malignant lesions of breast.

Eighty-five specimens of breast tissue were grouped and investigated by the technic of mixed cell agglutination reation (MCAR) for studying isoantigens A, B and H (O). MCAR was found to be strongly positive at the epithelium of acini in all the 25 subjects with benign lesions (10 fibroadenomas, 10 cystic mastitis and 5 gynecomastias). On the other hand, all the subjects with diagnosis of primary carcinomas in breast (45 subjects) and their metastases (15 subjects) were found to be uniformly negative for any agglutination reaction. In view of these findings, it seemed that the isoantigens in primary and metastatic malignancies of breast are always lost, while in the benign lesions isoantigens are always present. Although the exact mechanism is not understood, it seems that, in malignancies with a glandular differentiation, some differences exist when these are compared to malignancies with a squamous differentiation. The various possibilities are discussed in the light of previous studies.

Local immunoglobulin production in breast cancer.

Immunoglobulin levels (IgA, IgG, IgM) have been estimated in protein extracts of 55 malignant and 20 benign tumours of the breast, and in 17 normal tissues from a cancer bearing breast. IgA and IgG were significantly reduced in cancer compared with both benign and normal tissues but IgM, detected in only a third of tumours, was significantly increased. Total immunoglobulin levels and IgG in the malignant tumours correlated with plasma cell infiltration.The menstrual status of the patient had no influence on these findings.