Search

Accurate determination of HER2/neu status in breast carcinoma is essential. Alteration of preanalytic variables is known to affect HER2/neu results. American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) issued guidelines to standardize fixation for increased HER2/neu accuracy. We studied the effects of changing preanalytic variables on HER2/neu immunohistochemical and fluorescence in situ hybridization (FISH) results in a known HER2/neu+ invasive carcinoma. The clinical specimen was processed according to ASCO/CAP guidelines, with remaining tumor stored fresh without any fixatives for 4 days at 4°C and cut into core biopsy-sized pieces. Each was fixed in 10% formalin, 15% formalin, Pen-Fix (Richard-Allan Scientific, Kalamazoo, MI), Bouin solution, Sakura molecular fixative (Sakura Tissue-Tek Xpress, Torrance, CA), or zinc formalin for 0 to 168 hours. Immunohistochemical studies and FISH were performed. Compared with the clinical specimen, the samples showed no tumor degradation or marked difference by immunohistochemical studies, except the 1-hour 10% formalin and Bouin samples, or FISH, except the Bouin-fixed samples. Our study demonstrates that HER2/neu results remain accurate beyond ASCO/CAP-recommended preanalytic variables, with the exception of Bouin solution for FISH analysis.

KP-1 (CD68) is a recently described monoclonal antibody to a cytoplasmic epitope present on tissue histiocytes and macrophages. To determine the specificity and sensitivity of this marker in the evaluation of cases of malignant fibrous histiocytoma (MFH), this reagent and a panel of commercially antibodies were used to stain formalin-fixed paraffin sections from 25 cases of MFH and 25 other tumors, including a variety of soft-tissue sarcomas. Eighteen of 25 cases of MFH stained for KP-1 (72%), whereas all other tumors were negative, including 12 cases of pleomorphic soft-tissue sarcoma other than MFH. The percentage of tumor cells staining for KP-1 varied. In 11 cases KP-1 was only focally present, but staining was of a high intensity and associated with minimal nonspecific or background staining. Pleomorphic histiocytic cells and spindle cells from storiform tumors were strongly decorated with antibodies to KP-1 in most cases, and antigen also was present on tumor giant cells. Although alpha-1-antitrypsin and alpha-1-chymotrypsin stained a higher percentage of cases of MFH (92%), immunoreactivity for these markers also was noted in other tumors. Because of its specificity as a histiocyte marker, KP-1 is a useful component in a panel of antibodies for the characterization of soft-tissue sarcomas and the diagnosis of MFH.

Keratin profiles of exfoliated mesothelial and adenocarcinoma cells were determined using antisera to different molecular weight keratins (45, 46, 55, 63 kdaltons) and the immunoperoxidase technic. Most metastatic adenocarcinomas in effusions stained for low (45, 46 kdaltons) and intermediate (55 kdaltons) molecular weight keratins but were negative for 63 kdalton keratin. In contrast, most reactive and malignant mesothelial cells in effusions stained strongly for 63 kdalton keratin and keratins of lower molecular weight. This is the first report of high molecular weight (greater than 60 kdaltons) keratin in exfoliated cells of nonepidermal origin. Differences in staining for 63 kdalton keratin between mesothelial and adenocarcinoma cells may help to distinguish these cells in effusions.

This study describes an improved immunohistochemical method for the sensitive and specific identification of estrogen receptors (ERs) in paraffin sections from formalin-fixed and routinely processed breast carcinoma tissues, using DNase pretreatment to expose nuclear antigenic sites and commercially available immunoreagents (including monoclonal antibody) in kit form. Results were compared with dextran-coated charcoal cytosolic assay (DCC) and with conventional immunohistochemistry on frozen sections. Sensitivity and specificity for determinations on paraffin sections were 88% and 86%, respectively, and statistical analysis showed very good agreement between DCC and paraffin sections (kappa = 0.805). The DNase technic on paraffin sections allows excellent correlation between histologic characteristics and ER status and reduces DCC sampling error resulting from stromal dilution and tumor variability. This method offers a reliable and reproducible alternative when tissue is not suitable or unavailable for DCC or frozen tissue analysis and can be used for retrospective studies on stored tissue blocks.