Detection of c- erbB-2 activation in paraffin-embedded tissue by immunohistochemistry

Commercially available monoclonal antibodies were tested for their ability to detect increased levels of c- erbB-2 protein in formalin-fixed, paraffin-embedded breast carcinomas. Of five antibodies studied, four (TAB-250, CB11, 3B5, and N3/D10) showed strong cytoplasmic membrane reactivity in 23% (11 of 47) of routinely processed tumors, although interpretation of the immunoreactivity with 3B5 and N3/D10 occasionally was difficult due to cytoplasmic granular staining. Since the c- erbB-2 oncogene is activated by DNA amplification and overexpression of mRNA and protein, the same tumors were analyzed for c- erbB-2 activation by other techniques. c- erbB-2 activation in these 11 tumors was confirmed by immunohistochemistry of frozen tissue (nine of nine tumors), in situ hybridization (nine of 11 tumors), and Southern blot analysis (five of eight tumors). In some of these tumors the failure to demonstrate c- erbB-2 DNA amplification may be due to the small percentage of malignant cells. One additional tumor showed probable c- erbB-2 protein overproduction based on strong immunoreactivity with two antibodies (TAB-250 and CB11), although no definite activation could be demonstrated by additional techniques. Three other tumors (6%) showed equivocal c- erbB-2 protein overproduction based on weak immunoreactivity only with TAB-250, although unequivocal activation could not be demonstrated by additional techniques. The 32 carcinomas (68%) that showed no significant immunoreactivity with any antibodies in routinely processed tissue also showed no detectable c- erbB-2 activation by additional techniques. We conclude that TAB-250 and CB11 are reliable antibodies for detecting c- erbB-2 protein overproduction in routinely processed tissue. TAB-250 also weakly stains a few tumors showing no definite c- erbB-2 activation by other techniques. Two additional antibodies (3B5 and N3/D10) detect c- erbB-2 protein overproduction in paraffin-embedded tissue, but are more difficult to interpret. A fifth antibody, TA-1, is an excellent reagent for use on frozen tissue, but prolonged formalin fixation may impair recognition of its antigenic epitope.