Reliability of chromogenic in situ hybridization for detecting HER-2 gene status in breast cancer: comparison with fluorescence in situ hybridization and assessment of interobserver reproducibility.

Accurate determination of HER-2 status is important in the management of patients with breast cancer, especially in determining their eligibility for trastuzumab therapy. Fluorescence in situ hybridization (FISH) has been regarded as the gold standard method for detecting HER-2 gene amplification. Recently, chromogenic in situ hybridization (CISH), in which HER-2 is detected by a peroxidase reaction and the gene copies are determined by regular bright-field microscopy, has emerged as a potential alternative to FISH. However, this method requires validation before it can be adopted into clinical practice. In this study, we evaluated 80 cases of invasive breast carcinoma by CISH, compared the results with those obtained by FISH, and assessed interobserver reproducibility among three observers. We found that agreement among the three pathologists on the CISH-determined HER-2 status was achieved in 73 cases (91%), all of which had results matching the corresponding FISH results: 54 nonamplified and 19 amplified. Of the 19 amplified cases, 13 were scored unanimously as high-level amplification; six had a minor scoring discrepancy (ie, low-level vs high-level amplification). A major scoring discrepancy (ie, nonamplification vs amplification) was found in the remaining seven cases, three of which were amplified and four of which were nonamplified by FISH. Two of the latter cases had a polysomy of chromosome 17. The cases that caused scoring difficulty were those with an equivocal or borderline signal number against a high background. Overall, there was nearly perfect agreement between the CISH and corresponding FISH results, and interpretation of CISH results were highly reproducible among the three pathologists. We conclude that, in general, HER-2 status can be reliably assessed by CISH. Confirmatory FISH is recommended in cases with equivocal or borderline CISH copy numbers.