Critical evaluation of monoclonal antibody staining in breast carcinoma

The expression pattern of MUC1 (EMA) is related to tumour characteristics and clinical outcome of invasive ductal breast carcinoma

AIMS

To clarify MUC1 patterns in invasive ductal breast carcinoma and to relate them to clinicopathological parameters, coexpression of other biological markers and prognosis.

METHODS AND RESULTS

Samples from 243 consecutive patients with primary ductal carcinoma were incorporated into tissue microarrays (TMAs). Slides were stained for MUC1, oestrogen receptor (ER), progesterone receptor (PR), Her2/neu, p53 and cyclin D1. Apical membrane MUC1 expression was associated with smaller tumours (P = 0.001), lower tumour grades (P < 0.001), PR positivity (P = 0.003) and increased overall survival (OS; P = 0.030). Diffuse cytoplasmic MUC1 expression was associated with cyclin D1 positivity (P = 0.009) and increased relapse-free survival (RFS; P = 0.034). Negativity for MUC1 was associated with ER negativity (P = 0.004), PR negativity (P = 0.001) and cyclin D1 negativity (P = 0.009). In stepwise multivariate analysis MUC1 negativity was an independent predictor of both RFS [hazard ratio (HR) 3.5, 95% confidence interval (CI) 1.5, 8.5; P = 0.005] and OS (HR 14.7, 95% CI 4.9, 44.1; P < 0.001).

CONCLUSIONS

The expression pattern of MUC1 in invasive ductal breast carcinoma is related to tumour characteristics and clinical outcome. In addition, negative MUC1 expression is an independent risk factor for poor RFS and OS, besides 'classical' prognostic indicators.

Expression of mucins (MUC1, MUC2, MUC3, MUC4, MUC5AC and MUC6) and their prognostic significance in human breast cancer

Mucins are a large family of glycoproteins expressed by many epithelial cells and their malignant counterparts. Much interest has been focused on expression of its members in breast cancer because of their potential role as prognostic indicators and their involvement in cancer therapy. We have examined 1447 cases of invasive breast carcinoma with a long-term follow-up, using tissue microarray (TMA) technology and immunohistochemistry to evaluate the expression profiles of several mucins (MUC1, MUC2, MUC3, MUC4, MUC5AC and MUC6) and to assess their prognostic value. We detected MUC1 expression in 91% of tumours. MUC1 overexpression was associated with a lower grade, smaller tumour size, a higher oestrogen receptor (ER)-positive phenotype and absence of both regional recurrence and distance metastasis. The subcellular localization but not the level of expression had a prognostic value in predicting outcome. The aberrant cytoplasmic and membranous localization of MUC1 was associated with poor outcome compared with apical localization, which is the normal physiological site of expression. MUC2 expression was noticed in only 8.3% of all cases and was restricted to the cytoplasm of the tumour cells. An inverse trend was identified between MUC2 expression and lymph node stage and vascular invasion status. On excluding cases of mucinous carcinoma from the analysis, the inverse association with vascular invasion was still defined and in addition an inverse association with ER status emerged. MUC3 expression was detected in 91% of cases and its expression was associated with increased local recurrence, and lymph node stage. The membranous expression of MUC3 was found to be a potentially poor prognostic feature, with higher grade and poorer Nottingham Prognostic Index (NPI), and negative ER expression. MUC4, MUC5AC and MUC6 were expressed in 95, 37 and 20% of cases, respectively. Apart from an association between MUC4 expression and tumour grade and between MUC6 and ER-negative tumours, no other associations with any clinicopathological variables were found. Apart from the higher expression of MUC2 and MUC6 in mucinous carcinomas, no association was found between the expression of different mucins and tumour type. No association between the level of expression of any of the studied mucins and patient outcomes has been identified. In conclusion, most breast carcinomas express MUC1, MUC3 and MUC4. Among the various mucins expressed in breast cancer, MUC1 and MUC3 are potential prognostic indicators, MUC1 having the strongest relationship with patient outcome.

The importance of MUC1 cellular localization in patients with breast carcinoma: an immunohistologic study of 71 patients and review of the literature

BACKGROUND

The MUC1 mucin is present on the apical surface of normal secretory epithelia. In breast carcinoma, MUC1 expression is variable in amount and cellular localization, the significance of which is controversial. The authors undertook a detailed analysis of staining pattern combined with a comprehensive literature review to better understand the role of MUC1 in breast carcinoma.

METHODS

Seventy-one patients with breast carcinoma were examined for MUC1, beta-catenin, and E-cadherin staining patterns. These data were compared with data from 25 articles from the literature examining the expression of MUC1 in breast carcinoma.

RESULTS

All invasive carcinomas showed some MUC1 staining. In invasive ductal carcinomas, MUC1 was detected in the apical membrane (15%), cytoplasm (93%), or circumferential membrane (13%), with 81% of tumors showing a mixture of patterns. Tumors with low overall MUC1 expression (< or = 50% positive tumor cells) had a higher nuclear grade than tumors with high overall MUC1 expression (> 50%; P = 0.01). Tumors with high and low cytoplasmic expression had no difference in nuclear grade (P > 0.3). Circumferential membrane staining was correlated with positive lymph node status (P = 0.011).

CONCLUSIONS

In the literature, similar findings prevailed in which overall MUC1 expression was increased in lower grade (10 of 14 studies), estrogen receptor positive (8 of 13 studies) tumors and was associated with a better prognosis (8 of 13 studies). High cytoplasmic staining was associated with a worse prognosis, an association that was not explained by differences in histologic grade. Thus, the presence of MUC1 in the majority of tumor cells is associated with better differentiated tumors and with an improved prognosis. However, aberrantly localized MUC1 in the tumor cell cytoplasm or nonapical membrane is associated with a worse prognosis.

Immunohistochemical evaluation of canine carcinomas with monoclonal antibody B72.3

Monoclonal antibody (MAb) B72.3 reacts with a tumor-associated glycoprotein, designated TAG-72. TAG-72 has been identified in many human carcinomas but generally is not found in normal human tissues. Because of its proven utility in the diagnosis of human carcinomas, MAb B72.3 was applied to several different types of canine carcinomas. Five types of formalin-fixed, paraffin-embedded canine carcinomas were evaluated for immunoreactivity with MAb B72.3 by use of an avidin biotin immunoperoxidase complex method. Samples were considered positive when > or = 5% of all malignant cells contained a distinct intracellular stain. Immunoreactivity for MAb B72.3 was observed in 6/9 (67%) pulmonary adenocarcinomas, 7/13 (54%) transitional cell carcinomas, 7/11 (64%) mammary adenocarcinomas, 7/11 (64%) nasal adenocarcinomas, and 1/2 (50%) prostatic adenocarcinomas. The average cellular staining for positive carcinomas was 25%. Normal canine tissues from similar anatomic sites had little or no individual cell immunoreactivity. These preliminary results indicate that some canine carcinomas may express a tumor-associated antigen that is similar to TAG-72 and that MAb B72.3 immunoreactivity may be of diagnostic significance in classifying animal tumors.

The identification of informative parameters in the flow cytometric analysis of breast carcinoma

DNA ploidy and the measurement of proliferation or S-phase fraction are both of prognostic significance in breast cancer, yet clinical use is minimal in the U.K. Immunohistochemistry is, however, used to aid diagnosis, so a panel of antibodies were analysed by flow cytometry to assess their predictive value for prognosis, tumour stage and grade. Of 10 parameters tested on 202 breast tumour samples, tumour cell proliferation and DNA ploidy were the two most informative; cytokeratin staining, natural killer and B-cell infiltration also proved to be of value but there was no prognostic value in measuring tumour infiltrating monocytes, helper/suppressor T-cell ratios, tumour cell reactivity with carcinoembryonic antigen or human milk fat globulin antibodies. For each of the informative parameters, scores numerically weighted towards a poorer prognosis were derived which when combined, correlated with tumour grade, stage and prognosis. Such data interpretation is objective, and can be transposed to other human tumours.