We compared chromogenic in situ hybridization (CISH) with fluorescence in situ hybridization (FISH) for assessing HER-2/neu gene amplification using tissue microarrays (TMAs) made from formalin-fixed, paraffin-embedded tissue blocks from 113 cases of invasive breast carcinoma. TMAs were created using 0.6-mm tissue cores with 4 sampled cores per tumor. For both assays, a HER-2/chromosome 17 signal ratio of 2.0 or more was considered positive for gene amplification. The average ratio of cores from the same tumor was used for determination of gene amplification status of that particular tumor Of 113 cases, 102 were tested successfully by both assays. The results were concordant in 100.0% of cases (63 amplified; 39 nonamplified). All 22 cases of borderline (ratio, 2.0-2.5) or low-level (ratio, 2.6-3.9) amplification by FISH also showed HER-2 gene amplification by CISH. CISH is as sensitive as FISH in detecting borderline and low-level HER-2 amplification. Reliable recognition of the invasive carcinoma area by light microscopy and preservation of the test slides are added advantages of CISH. CISH performs as well as FISH in the analysis of HER-2 gene amplification in breast cancer and might have advantages in certain situations.