Development and validation of a sensitive immunohistochemical oestrogen receptor assay for use on archival breast cancer tissue

Concordance of the hormone receptors and correlation of HER-2/neu overexpression of the metachronous cancers of contralateral breasts

The primary objective of this study was to evaluate the relative prevalence of estrogen receptor-negative contralateral breast cancer to the first primary cancer and to assess the correlation between the relative overexpression of HER-2/neu in the first primary cancer and contralateral breast cancer. A total of 144 women diagnosed with cancers in contralateral breasts were identified from the Henry Ford Health System tumor registry. Data were retrieved from electronic databases and medical records. Women were dichotomized into users and nonusers of tamoxifen. Hormone receptors were scored as positive or negative. HER-2/neu overexpression, assessed by immunohistochemistry, was scored as 0, 1(+), 2(+), or 3(+). Concordance between hormone receptors of the two cancers was low (kappa = 0.27, p = 0.06). Stratification of women by tamoxifen therapy yielded an almost fivefold increase in the proportion of estrogen receptor-negative cancers among the users, while the proportion of cancers expressing no estrogen receptor remained the same among the nonusers (39.6% versus 40.6%). Matched, archived, paraffin-embedded specimens of the first and contralateral breast cancers were available for 57 women. The correlation between the relative overexpression of HER-2/neu between the first primary and the contralateral breast cancer was 0.4 (p = 0.002). The higher prevalence of estrogen receptor-negative contralateral breast cancer among tamoxifen users concurs with previous reports. The biological mechanism for this observation is not understood; however, it has been proposed that tamoxifen inhibits the proliferation of estrogen receptor-positive breast cancer cells, while estrogen receptor-negative cells may continue to grow because of selective pressure. The correlation between HER-2/neu overexpression in the matched first primary and contralateral breast cancers was statistically significant, suggesting that the diagnosis of HER-2/neu overexpression in contralateral breast cancer is associated with HER-2/neu overexpression in the first primary cancer.

Predicting aggressive outcome in T1N0M0 breast cancer

Despite the excellent overall prognosis, unpredictable breast cancer recurrences and deaths also occur among T1N0M0 patients. We have evaluated clinically applicable methods for identifying aggressive outcome in T1N0M0 breast cancer. The material is based on aggressive T1N0M0 invasive ductal and lobular carcinomas diagnosed in Turku University Hospital and Jyväskylä Central Hospital, Finland, during 1987-1997. We studied all the T1N0M0 breast cancers that had led to recurrency or death (n=21, 95% T1cN0M0) during the follow-up period (4-14 years). The study is based on statistical analyses of matched case-control data in which the prognostic factors of each individual patient with aggressive disease were compared with control patients (n=45) individually matched by tumour size, age at diagnosis, histological type of tumour and length of follow-up. The cancer cases were examined for clinically applicable conventional and immunohistochemical pathologic prognostic factors. High Ki-67 immunopositivity was the strongest prognosticator of breast cancer death or recurrence in T1N0M0 breast cancer. Also, high p53 immunopositivity, low oestrogen receptor immunopositivity and Her-2/neu oncogene amplification by chromogen in situ hybridisation were reliable indicators of unfavourable outcome. Our statistical methods also allowed us to determine for the present material a range of clinical significance for each immunohistochemical prognostic feature with the associated relative risk for breast cancer death and recurrence. The paper suggests guidelines for predicting aggressive outcome in T1N0M0 breast cancer.

A new immunohistochemical antibody for the assessment of estrogen receptor status on routine formalin-fixed tissue samples

We describe a method for the immunocytochemical assessment of estrogen receptor (ER) status on routinely processed, formalin-fixed tissue using a recently developed commercially available monoclonal antibody (Dako 1D5) with a microwave antigen retrieval technique. A series of 90 cases of human breast carcinoma was analyzed and the staining was assessed using a semiquantitative microscopic scoring method and by assessment of the percentage of nuclei showing positive staining. The results were compared with assessment using another commercially available antibody (Abbott H222) and with clinical response to tamoxifen therapy. Direct comparison of the paired sets of H scores and the percentage of positively stained nuclei using the two different techniques showed a significant correlation. By assigning an arbitrary cut-off for positivity of H score = 50, assessment of ER status using DAKO 1D5 antibody was found to correlate with response to tamoxifen therapy with a sensitivity of 90% and a specificity of 51%, similar to values in previous studies using other methods. We conclude that assessment of ER status using this new antibody and technique gives accurate results on routinely processed, formalin-fixed tissue and may be used as an alternative to other methods.

Classical oestrogen receptor is not detectable in pancreatic adenocarcinoma

Recent reports have suggested the presence of oestrogen receptors (ER) in pancreatic carcinoma. Therefore the tumour may be sensitive to hormone manipulation therapy. We examined 23 biopsies of human pancreatic carcinoma tissue for the presence of ER. The tissue was assayed by two methods: Iso-electric focusing (IEF) and ER-ICA an immunocytochemical assay. All biopsies were tested with ER-ICA and ten by IEF. Each biopsy was assessed histologically for tumour content and 20 contained adenocarcinoma. None of the samples of pancreatic carcinoma were positive for ER using the ER-ICA, and none demonstrated the binding peak typical of ER in the IEF assay. These results suggest that in pancreatic carcinoma oestrogen receptor is either absent (or at very low levels), or of a different type to that found in breast and uterine tissue. On theoretical grounds at least, this must raise questions as to the oestrogen sensitivity of pancreatic cancer.

The effect of estrogen on the estrogen receptor-immunoreactive cells in the rat medial preoptic nucleus

The localization of estrogen receptor-immunoreactive (ER-IR) cells in the medial preoptic nucleus (MPN) and the effect of estrogen upon these cells were quantitatively analyzed by immunohistochemistry using a monoclonal antibody to estrogen receptor (ER) protein in rat. ER was distributed in the nucleus of MPN neurons and was not detected in either cytoplasm or glial cells. There were more ER-IR cells with higher immunoreactivities in ovariectomized (OVX) rats than in estradiol (E2)-treated rats. The number of ER-IR cells in E2-treated compared with OVX rats was reduced by 43%. In particular, the number of ER-IR cells in the central part of the MPN was largely decreased. Our data indicate that the central part of the MPN is most sensitive to estrogen.

Ultrastructural localisation of oestrogen receptor in breast cancer cell nuclei

A method of localising oestrogen receptor in nuclei of breast cancer cells by the protein A-gold technique is described. Mast cell granules were found to take up uncoated particles of the gold sol selectively.

Oestrogen receptor in breast cancer: prognostic studies using a new immunohistochemical assay

A recently developed and validated histochemical method--immunogold-streptavidin enhancement--was used to determine oestrogen receptor content in paraffin sections of breast cancers. The selection of the best cut-off point to define oestrogen receptor status as rich or poor was made on the basis of survival data, using the following indices: survival to term; disease-free interval; survival at 5 years; and disease-free interval at 5 years. Oestrogen receptor status was examined in relation to histological grade, lymph node status, menopausal status and tumour size and these four indices were considered as independent prognostic factors. Semi-quantitative assay of receptor content showed that increasing content was related to better prognosis. Adjuvant therapy alone had no effect on patient outcome. Independently, histological grade and lymph node status, but not menopausal status or tumour size, were prognostic indicators. Oestrogen receptor rich status, as measured by immunogold-streptavidin, in conjunction with certain of these factors indicated a better prognosis. This was comparable with results in reports using other methods of receptor assay. We found the oestrogen receptor status and menopausal status more significant at 5 years than at term. The advantage of immunogold-streptavidin enhancement, which we found as reliable as other methods for oestrogen receptor assay, is that it can be used on archival, routinely paraffin-processed material.