Fluorescence in situ hybridization (FISH) has both excellent sensitivity and specificity in detecting HER2 gene amplification in invasive breast carcinoma. FISH has not been widely implemented in clinical practice because of reagent costs and the special instrumentation and expertise required to perform and integrate the assay. Immunohistochemistry (IHC) for HER2 protein is widely used, but false-positive and false-negative results are problematic. We developed a bright-field assay to visualize HER2 gene amplification and concomitant HER2 protein expression (EnzMet GenePro). This assay detects HER2 gene amplification via deposition of metallic silver by enzyme metallographytrade mark (EnzMettrade mark, Nanoprobes, Yaphank, NY) combined with HER2 protein detection by IHC using alkaline phosphatase and fast red K substrate visualization (CB11;Ventana, Tucson, AZ). The assay was performed on 94 invasive breast carcinomas, for which FISH (PathVysiontrade mark, Vysis, Downer's Grove, IL), conventional IHC (CB11), and enzyme metallography (EnzMettrade mark) results were known. The EnzMettrade mark component of the assay was scored as either HER2 gene amplified, polysomic, or nonamplified. The IHC component was scored using the conventional FDA scale of 0 to 3+. Concordance of the EnzMet component of the assay versus FISH was assessed and showed an excellent correlation (Pearson coefficient of 0.95; P < 0.001). The combination of gene and protein detection (EnzMet GenePro) displayed a specificity of 100% and an accuracy of 92.6% (95% confidence interval 85.3-97.0), facilitated recognition of gene/protein discordances, and allowed for efficient interpretation of the slide by conventional light microscopy. The interobserver kappa for each component was excellent (IHC, kappa = 0.94; and EnzMettrade mark, kappa = 0.96). EnzMet is the first bright-field ISH assay in our experience that routinely and nonambiguously detects endogenous HER2 signals, essential for a reliable clinical HER2 assay, and in combination with HER2 protein enables improved diagnosis in borderline cases.