The American Society of Clinical Oncologists and College of American Pathologists have recently released new guidelines for laboratory testing of HER2 status in breast cancer, which require high levels (95%) of concordance between immunohistochemistry positive (3+) and fluorescence in situ hybridization-amplified cases, and between immunohistochemistry negative (0/1+) and fluorescence in situ hybridization-nonamplified cases; these required levels of concordance are significantly higher than those found in most published studies. We tested the hypothesis that a modification of the HER2 immunohistochemistry scoring system could significantly improve immunohistochemistry and fluorescence in situ hybridization concordance. A total of 6604 breast cancer specimens were evaluated for HER2 status by both immunohistochemistry and fluorescence in situ hybridization using standard methodologies. Results were compared when the standard immunohistochemistry scoring system was replaced by a normalized scoring system in which the HER2 score was derived by subtracting the score on the non-neoplastic breast epithelium from that on the tumor cells. Among the 6604 tumors, using a non-normalized immunohistochemistry scoring system, 267/872 (30.6%) of the immunohistochemistry 3+ cases proved to be fluorescence in situ hybridization nonamplified, whereas using the normalized scoring system only 30/562 (5.3%) of immunohistochemistry 3+ cases proved to be 'false positive'. The concordance rate between immunohistochemistry 3+ and fluorescence in situ hybridization-amplified cases using the normalized scoring method was 94.7%, whereas the concordance using the non-normalized method was only 69.4%. Extremely high concordance between immunohistochemistry and fluorescence in situ hybridization assessment of HER2 status in breast cancer is achievable, but to attain this high level of concordance, modification of the FDA-approved immunohistochemistry scoring system is required.