We compared the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) immunohistochemical scoring criterion (30%) for determining HER2 status and the Food and Drug Administration criterion (10%) with fluorescence in situ hybridization (FISH), the HER2 gene amplification method in 328 cases of breast cancer. Of 294 tumor samples successfully analyzed simultaneously by FISH and immunohistochemically, 178 of 196 cases scored 3+ using the 10% and the 30% criteria. Using FISH as the reference, the number of false-positives was reduced from 24 to 9 after application of the 30% criterion. The specificity of immunohistochemical analysis was higher with the 30% (92.0%) vs the 10% (78.8%) criterion. The kappa coefficient between FISH and immunohistochemical analysis was increased to 0.850 (almost perfect agreement; P < .001) after application of the 30% criterion vs 0.757 (substantial agreement) for the 10% criterion; the false-positive rate decreased to 5.1% from 12.2%. The chi(2) test showed that immunohistochemical analysis had significantly higher accuracy with the 30% (94.9%) vs the 10% (87.8%; P = .014) criterion. Our results from a large series of Chinese patients with breast cancer support that the ASCO/CAP 30% criterion may offer better results for assessing HER2 status.